Condition dependent strategies of egg size variation in the Common Eider Somateria mollissima.

We analysed intraclutch egg-size variation over the laying sequence in relation to clutch size, and the relation between clutch size and female body condition, in the Common Eider Somateria mollissima during an 8-year period. The aim was to assess if eiders adjusted egg size within the laying sequence depending on clutch sizes in response to body condition, as such an adjustment could have adaptive implications on reproductive success through a size advantage for the hatchlings. The analyses were performed on a population level; and then at the individual level using data from recaptured females that changed clutch size between years. Based on 1,099 clutches from 812 individual females, population clutch size averaged 4.13 eggs (range: 1-6), with 4- and 5-egg clutchesconstituting c.70% of all clutches, taking turns in being the most represented clutch size. Clutch size was positively related to female pre-laying body condition at both the population and individual levels. Egg size varied significantly within and between clutch sizes and changes were significantly related to the laying sequence. First eggs were significantly larger in 4-egg clutches and second eggs marginally smaller than in 5-egg clutches, a pattern also found among individual females changing clutch size between years. The relationship between female pre-laying body condition and clutch size, and the intraclutch egg-size pattern indicate that both clutch size and egg size are actively adapted to the pre-breeding body condition of the female. As egg size potentially optimise reproductive success through a size advantage in hatchlings, the observed pattern of intraclutch egg-size variation suggests that female eiders possesses a finely tuned conditional dependent mechanism that may optimize reproductive output in years were females are in suboptimal body condition for breeding.

Introduction

Intraclutch egg-size variation has frequently been described in both altricial and precocial avian species [1-3]. In large long-lived precocial waterfowl species, such as geese and eiders, egg-size variation over the laying sequence generally follow a rather fixed curvilinear pattern, with increasing size from first to second and/or third egg and then a declining size over the following eggs in the laying sequence, with the final egg being the smallest [4-10]. With few exceptions, this pattern of intraclutch egg-size variation is consistently found among the most commonly laid clutch sizes within different eider populations, occurring almost independently of clutch size, female age and condition, and time of laying [3,4,7,11,12]. Parsons [13] suggested that the small size of the first egg was related to a lower physiological efficiency of females at the up-start of egg laying, while other studies support a physiological and hormonal, genetically based, explanation to the overall pattern of intraclutch egg-size variation [see 4]. Alternative but not mutually exclusive hypotheses, have related intraclutch egg-size variation to the depletion of resources in laying females with egg number, to incubation strategy, to facilitation of hatching synchrony and to differential investment in eggs with highest probability of hatching [8,10,11,14-22]. Despite the long-term focus on intraclutch egg-size variation, neither adaptive nor non-adaptive explanations have proved to fully explain the observed egg size patterns.

Although egg size in birds generally show high repeatability and heritability, and consequently is less flexible in individual birds than e.g., clutch size [cf. 23–25], egg size and intraclutch egg-size variation may have important implications for the reproductive success of individual females. In geese and eiders, wich lay small clutches of 4 to 6 (1–7) relatively large eggs, larger eggs within the clutch produce larger hatchlings that have a faster growth rate, higher pre-fledging survival and even a higher recruitment rate [3,26–31,but see 32]. However, despite the apparent benefit of making adaptive changes (e.g. laying larger eggs when laying a reduced clutch size or vice versa, or changing egg size according to female conditional state or in relation to environmental conditions), such adaptive intraspecific changes have not been substantially documented [see 5,33–35]. This is surprising, as even small proportional changes in egg size that would make equal egg size within clutches, is expected not to be costly to females already being physiologically upgraded for egg laying [cf. 11,24]. Likewise, avian life history characteristics suggests that adaptive patterns in both clutch and egg size should be more prevalent in species that store resources for reproduction in the pre-breeding period (capital breeders) compared to species depending on resources available during the laying and chick-rearing period (income breeders) [36]. Thus the apparent lack of flexibility in egg size even in capital breeding birds, suggests that adaptations to varying conditional states of females is mainly adjusted through changes in clutch size [cf. 24]. Indeed, many studies have shown a proximate linkage between female body condition and clutch size, as well as an increase in reproductive success with increasing clutch size [e.g., 14,23,37–39].

In a long-term study of breeding eiders in Denmark, records of egg size showed that intraclutch egg-size variation over the laying sequence followed the common curvilinear pattern as generally described in eiders [see 7,9]. However, we also observed that egg size in the most commonly laid clutches of 4 and 5 eggs showed a systematic difference with respect to the size of the first laid egg. Examining previous studies of eider clutch and egg size [cf. 7,9,10], we found a similar difference in other breeding populations, suggesting that eiders adaptively change the size of first laid egg depending on clutch size.

In the present study, we use data from a long-term survey of breeding eiders to test if the common eider females change egg size within the laying sequence depending on clutch size, which may have adaptive value in terms of reproductive success. Since clutch size in the eider is closely related to female body condition, we also included female body condition in analyses of clutch size and clutch size changes. We analyse data at the population level, as well as the individual level, based on ringed and recaptured individual females that changed clutch size within our study period of 8 years, and discuss our results in relation to the potential impact on reproductive success.

Materials and methods

Data collection

Data were collected annually during 1993–2000 at the island of Saltholm, Denmark (55°55’N 12°46’E). Access permits to the island was provided by the Saltholm Ejerlaug and the Danish Forest and Nature Agency. During the study period the island, covering 13.5 km2, held an estimated breeding population of 4,000 to 6,000 eiders [40]. In all years, seven 45 m wide coast-to-coast transects, covering an area of c. 0.75 km2, were searched thoroughly for nests towards the end of the incubation period in early May. Transects were distributed evenly 1 km apart following east-west oriented UTM grid lines. Attempts were made to capture all encountered incubating females, but only data from clutches where females were caught (assumed incubating on a fully laid clutch) were included in the present analyses. Two females sitting on clutches of 7 eggs were omitted as egg size and coloration strongly indicated egg dumping in both nests. We consider that the extremely low nest densities (average 0.65 nest per 45 x 45m sampling unit ~ 3.2 nest per ha [40]) relative to other studies generally precludes nest parasitism, which increases with nest density, being reported in colonies above 10–15 nest per ha [cf. 41]. However, we cannot reject that a few incomplete or parasitized nests, as well as some partially predated nests (in cases where no egg-shell remains was found to disqualify the record) may unintentionally have been included in the present data, and hence have induced some uncontrolled variation.

During 1993 to 2000, 1,099 incubating females were captured on nests (between 79 and 174 per year), ringed with standard steel leg bands and body mass (to nearest 5 gram) and tarsus length (to nearest 1 mm) were recorded. The sample included 812 individual females, of which 610 were caught once, 141 twice, 44 thrice, 12 four times, 3 five times and 2 six times.

Captures and ringing of eiders were conducted according to the practical and ethical guidelines for bird ringing and carried out under the institutional ringing license (no. 600) issued by the Ringing Centre at the Natural History Museum of Denmark.

The 1,099 clutches contained a total of 4,531 eggs. In all clutches, individual eggs were ordered in their laying sequence (first to sixth egg) based on individual degree of staining [4,8,34, see also 42], a method previously applied to eiders [23]. Egg length and width were measured to nearest 0.1 mm using callipers.

Egg volume (hereafter referred to as egg size) was subsequently calculated according to the formula: Volume (egg size) = length × π × bredth2 × 0.00164 cm3 [9,43]. In each year, female body mass prior to egg laying (Wstart) was estimated for each individual female based on the difference between body mass at capture (Wcapt) and the average year specific body mass at hatching (Whatch). Whatch was obtained from regressions of body mass on the tarsus length (cubed) of all females captured with ducklings or hatching eggs in their nests. This calculation was done separately for all years, and hence accounts for annual differences in average population body condition (weight related to individual structural size). We used a daily weight loss of 20 gram during incubation and an incubation period of 28 days [cf. 44] to calculate individual body condition at start of egg-laying as where Δw is the difference between Wcapt and Whatch for a given structural size. Clutch mass was estimated setting average egg mass to 100 gram [cf. 45,46].

To ensure that clutch size differences were not responsible for differences in Wstart-estimates, we tested relations between clutch sizes and observed Whatch. The analysis suggested that clutch size did not affect Whatch whereas tarsus3 significantly related to Whatch (see S1 File).

In the analyses of intraclutch egg-size variation, we only included clutches of 2 to 6 eggs (98.5% of all clutches recorded) given the uncertainty associated with clutches of 1 (likely incomplete) and 7 eggs (potentially parasitized) (please see S1 Table for distribution of clutch sizes). When analysing individual egg-laying patterns in recaptured individuals, we only included 4 and 5-egg clutches, as these were numerically dominating (> 70% in all years).

Statistics

Clutch size

We used general linear models (GLM) to test if clutch size differed between years and whether clutch size related to female body condition at the start of egg laying. To test whether clutch sizes of four or five dominated in different years we used a chi–square test [47].

Clutch size and body condition for recaptured females

For recaptured females, the likelihood of changing their reproductive strategy with regard to clutch size in a given year was tested by relating between-year changes in clutch size to between-year changes in body mass at the start of egg laying. We classified female reproductive strategy as either reducing, maintaining or increasing clutch size. To account for multiple contributions by some individuals (up to 4 recaptures in different years) we included individual females as a random factor in a mixed model. The likelihood of a change (increasing, maintaining or reducing) in clutch size as a response to differences in pre-laying body mass were tested by a generalized linear model with a multinomial distribution. We calculated least square pairwise differences for post hoc comparison of differences in egg sizes between clutch sizes. The proportion of significant pair-wise differences by far exceeded 5% expected by random, and hence did not require adjustment of significance threshold.

Inter- and intraclutch egg-size variation

We tested egg size differences within and between clutch sizes for first, second and third egg, separately using a mixed model including clutch size as a fixed effect and year as a random factor.

Patterns of egg size variation in relation to laying order

We tested the pattern of egg size differences over the laying order among clutch sizes using a repeated measure ANOVA, to detect if the pattern with laying order differed between different clutch sizes. The model included laying order, clutch size and year, and interactions between laying order and clutch size, and between year and laying order. In this model, we used clutch size and laying order as discrete variables, because this bypasses the need to specify the relation (linear or non-linear). The model used individual clutches as the repeated measure. Note that the relationship between laying order and egg size could only be compared for the number of eggs that corresponds to the smallest clutch size selected (e.g. the pattern for the first four eggs could only be tested in clutches with at least four eggs). We therefore conducted separate repeated measures ANOVA for the first 2 eggs, then on the first 3 eggs, then on the first 4 eggs and then on the first 5 eggs.

Egg-laying patterns within individuals

The recaptured females allowed us to test if the strategy for 4 and 5 egg clutches differed when individual variation was removed. We used paired t-tests to test if the egg size of first and second egg differed in 4 and 5 egg clutches. To account for changes in egg strategy we analysed females that changed clutch size from 4 to 5 eggs separately from females that changed clutch size from 5 to 4 eggs. Eight females, however, were recaptured three times and thus entered the data set for both changes from 4 to 5 and from 5 to 4 egg clutch sizes.

Assumptions regarding normal distribution and homoscedasticity of residuals were fulfilled for all tests requiring it. We used SAS 9.4 (SAS Institute, Cary, NC) for all analyses using Proc Mixed, Proc GLM, Proc Ttest, Proc Univariate, and Proc Freq.

Results

Clutch size

Average clutch size during 1993–2000 was 4.13 ± 0.03 SE (N = 1,099). Clutch size approached statistically significant variation between years, with the largest clutch sizes recorded in 1993 and 1996 and the smallest clutch size in 1997 (range: 3.97–4.30; General linear model: F7,1091 = 1.99, p = 0.054, R2 = 0.013)(Fig 1) (S2 Table).

Fig 1

Average clutch size (±SE: Standard error) for 1,099 clutches of eiders captured during incubation on the island of Saltholm during 1993–2000.

In all years, clutches of 4 and 5 eggs were the most frequently occurring clutch sizes (annual average 35.9%, range: 24.5–47.3% and 37.8% range: 30.3–50.4% respectively). A significant (Chi-square test: χ62 = 20.2, p < 0.005) shift in dominance occurred with 4 egg clutches being more frequent in some years and 5 egg clutches in other years. On average total clutch volume of 4-egg clutches was 80.3% of the volume of 5-egg clutches.

Clutch size was significantly and positively related to female body condition at the start of egg-laying (General linear model: clutch size: F1, 809 = 1382.53, p < 0.0001, slope = 0.006).

Clutch size and body condition for recaptured females

The majority of the recaptured females that reduced clutch size between years, showed a decrease in body mass (72 lost weight and 18 gained weight). The majority of the recaptured females that increased clutch size between years, showed an increase in body mass (86 gained weight and 12 lost weight). Among the 108 females that did not change clutch size between years there was no distinct pattern with regard to changes in body mass (43 females lost weight and 65 gained weight). The changes in individual female body mass at start of incubation between years significantly affected changes in clutch size (increasing, maintaining or reducing clutch size)(Generalized linear mixed model F1, 287 = 104.6, p < 0.0001, slope = 0.011). Hence, when females lost body mass relative to the previous breeding record, they were significantly more likely to reduce clutch size, as indicated by the model estimates, and vice versa.

Inter- and intraclutch egg-size variation

Overall mean egg size in all clutches (N = 4,531) was 103.1 ± 0.1 cm3. Excluding extreme outliers of one dwarf (20.1 cm3) and one extremely large egg (176.7 cm3), egg size range was 79.4–159.5 cm3 (n = 4,529) and the smallest egg was 49% the size of the largest egg in the population of measurements.

Egg size showed a consistent pattern with laying order across different clutch sizes (Fig 2), and the size of the first, second and third egg differed significantly with clutch size (Table 1). The difference in the size of the first egg was related to a significantly larger egg in 4-egg clutches compared to all other clutch sizes (Fig 2, S3 Table). The second egg in the laying sequence was significantly larger in 5-egg clutches compared to 2- and 3-egg clutches, almost significantly larger (p = 0.054) than in 4-egg clutches, but did not differ from the second egg in 6-egg clutches (Fig 2) (S3 Table). The size of the third egg in 4-egg clutches was significantly larger than in 3-egg clutches, but significantly smaller than in 5- and 6-egg clutches.

Fig 2

Average egg size (±SE: Standard error) in eider clutches in relation to position in the laying sequence for females captured on nests with 1 to 6 eggs.

Table 1

Test of differences in egg size of first, second and third eggs across clutches of 2–6 eggs.

Egg size	df	F	p	N
First egg	5, 804	5.49	<0.001	817
Second egg	4, 796	13.87	<0.001	808
Third egg	3, 752	17.22	<0.001	764

Data were analyzed with a mixed model with year as a random factor and only included first records of recaptured females. Bold font indicates statistically differences in egg size.

Patterns of egg size in relation to laying order

Egg size was significantly affected by laying order in clutch sizes of 2–6 eggs (Table 2). More importantly, clutch size significantly affected the size of eggs within the laying sequence as indicated by the significant interactions between laying order and clutch size for clutch sizes between 2 and 6 eggs (Table 2).

Table 2

Test (repeated measures ANOVA) of egg size changes with clutch size, year and laying order, and the interaction terms for laying order for a stepwise increase in clutch size (ex: Egg 1 and 2 includes all clutch sizes larger than one egg (2 to 6 egg clutches); Egg 1 to 5 includes clutches with 5 or more eggs (5 and 6 egg clutches)).

	Egg 1 and 2 (Nclutch = 1,087)			Egg 1 to 3 (Nclutch = 1,030)			Egg 1 to 4 (Nclutch = 844)			Egg 1 to 5 (Nclutch = 450)			Egg 1 to 6 (Nclutch = 35)
	df	F	p	df	F	p	Df	F	P	df	F	p	df	F	p
clutch size	4, 1073	10.05	<0.0001	4, 1017	11.86	<0.0001	3, 832	4.87	0.008	2, 439	1.32	0.251
year	7, 1073	0.99	0.437	7, 1017	1.24	0.276	7, 832	1.00	0.430	7, 439	0.66	0.705	6, 26	0.83	0.554
order	1, 1073	16.42	0.024	2, 2034	21.03	<0.0001	3, 2496	59.16	<0.0001	4, 1756	49.73	<0.0001	5, 130	13.70	<0.0001
order*clutch size	4, 1073	16.14	<0.0001	8, 2034	18.63	<0.0001	9, 2496	16.78	<0.0001	8, 1756	4.49	0.001
order*year	7, 1073	1.71	0.102	14, 2034	0.96	0.491	21, 2496	0.97	0.495	28, 1756	0.99	0.481	30, 130	0.95	0.552

Bold fonts indicate statistically significant differences.

Egg laying patterns within individuals

The difference in the intraclutch egg-size pattern of females laying 4 and 5 egg clutches in the general analyses (cf. Fig 2), was also found in recaptured females that changed their clutch size from 4 to 5 eggs (N = 35)(Fig 3A) and from 5 to 4 eggs (N = 43)(Fig 3B). When laying a 4-egg clutch, the size of first and second eggs were comparable in size (Paired t-test t34, p = 0.839), but there was a marked size difference (Paired t-test t34 = 3.78, p = 0.0006) when subsequently laying a 5-egg clutch, relating to a reduced size in the first egg and a slight increase in size of the second egg. For females changing clutch size from 5 to 4 eggs, the opposite change in the size of the first and second egg was evident (Paired t-test: 5-egg clutch: t42 = 4.432, p<0.001; 4-egg clutch: t42 = 0.07, p = 0.943t). Hence, individual females laying a 4-egg clutch increased the size of the first egg while decreasing the relative size of the second egg compared to years when laying a 5-egg clutch and vice versa. A similar, but less clear, trend was observed among females making two changes, either 4-5-4 eggs or 5-4-5 eggs, although the samples sizes were very small (N = 5 and 6 respectively) and hence subject to differences between individuals (Fig 4).

Fig 3

Average egg size (±SE: Standard error) for consecutive eggs in the laying sequence for individual female eiders that changed clutch size from A) 4 to 5 eggs (N = 35) and B) 5 to 4 eggs (N = 46).

Fig 4

Average egg size (±SE: Standard error)) for egg number 1 and 2 in the laying sequence for individual female eiders that made two clutch size changes: From 4 to 5 to 4 eggs (N = 5) and from 5 to 4 to 5 eggs (N = 6) between years.

Discussion

Our results confirmed a general intraclutch increase in egg size from the first to the second egg, with declining size in subsequent eggs, as commonly described in eiders and in several other Anatidae species [4,7,9,48,49]. However, we also found a consistent pattern of intraclutch egg-size variation depending on clutch size for the most commonly laid clutches of 4 and 5 eggs. In 4-egg clutches, the first egg was significantly larger and the second egg almost significantly smaller than in 5-egg clutches. This population level pattern in egg-size with clutch size was also found among individual females when they both increased and decreased their clutch size between years. As clutch size was significantly related to female body condition, our results suggest that intraclutch egg-size variation may be an adaptive strategy adopted by individual females in response to their own pre-laying body condition.

Clutch size variation

Female eider body condition prior to breeding is widely acknowledged as a major determinant of clutch size [23,50,51]. However, clutch size is also affected by female age and timing of laying [52,53], which often are interrelated, with young birds laying later than older birds [54]. In this study, nest surveys were always performed in the early hatching period, and annual average laying date varied by only five days throughout the study period [40], suggesting no effect of year from the timing of laying. Likewise, restricting the analysis to females caught on nests minimised the possible inclusion of late breeding young birds [cf. 55], as well as potential re-nesters. We therefore believe that the variation in average clutch size between years mainly reflects varying environmental conditions affecting females during winter and the pre-breeding period.

Intraclutch egg-size variation

Overall, the present data reflect the general pattern of intraclutch egg-size variation reported in many waterfowl species as well as in other eider populations [e.g., 7,9,11]. However, we show that egg size was significantly related to laying order (in 2–6 egg clutches), and, for the first time, that the pattern of egg size variation with laying order was significantly affected by clutch size (in 2–6 egg clutches). This suggests that eider females produce eggs with a specific predetermined pattern of intraclutch size variation depending on ultimate clutch size. Since this pattern alternated in individual females when both increasing and decreasing clutch size between years (in relation to associated pre-laying body mass), our results strongly support the interpretation that eider females actively adapt clutch and egg size as part of their reproductive strategy, following some pre-programmed clutch/egg size “formats”.

Overall, the absolute difference in the size of the first and second eggs in 4 and 5-egg clutches was ~1.4 cm3, corresponding to approx. 1.2% of the average egg size of the first two eggs. This relatively small change in the first laid egg has not been commented in previous studies, even though matching patterns showing a larger first egg in 4-egg clutches compared to 5-egg clutches, have been reported in other eider populations (Fig 7 in [7]; Table 1 in [9], see also [10]).

Implications of changing egg size

Egg size is recognised as a major determinant of hatchling success, and increases in egg size directly affects the size of ducklings within and between clutches, and hence may have significant fitness value [56,57,]. Likewise, reproductive success, in terms of number of ducklings leaving the nests and higher duckling survival, increases with clutch size [58-60]. Hence, increasing the size of the first laid egg in 4-egg clutches, has the potential to partly compensate for overall reduced reproductive success compared to a higher success in 5 and 6 egg clutches. Thus, when laying a 4-egg clutch, females produce two large eggs, which are almost comparable in size with the two largest eggs in a 6-egg clutch and comparable to the second largest egg in a 5-egg clutch (cf. Fig 2). Hence, instead of producing only one large egg, if the size of the first egg was similar to that of 5 and 6 egg clutches, females laying 4-egg clutches lay two large eggs and hatch two relatively large ducklings, with the likely associated fitness benefits. In geese and large waterfowl species, greater size at hatching is reported to benefit individuals by a stronger competitive ability for optimal brooding, more efficient foraging skills, stronger and quicker responses to predator warning in female and/or a more efficient physiological/immunological state. Such traits likely contribute to an increase in survival [cf. 28,61].

The advantage of laying two relatively large eggs may however, not only be restricted to the benefit of larger size in individual ducklings within a clutch of 4 eggs. As eiders frequently engage in brood amalgamation, forming so-called crèches [62,63], larger ducklings may benefit from their size advantage at the multi-brood/cohort level when grouped with other ducklings from one or more broods, or abandoned by the female into crèches [but see 64]. Analysed in relation to the expected size variation in a duckling cohort based on egg sizes, the present data (cf. Fig 2) show that among the largest 20% of all ducklings in an average cohort, almost half (47.6%) originate from 4-egg clutches (from the first and second egg), 48.6% originate from 5-egg clutches (from the second and third egg), and 3.8% from 6 egg clutches (from the second and third egg). Without the recorded size increase in the first egg (compared to the size of first egg of 5- and 6-egg clutches), 4-egg clutches would only represent 31.2% of the largest ducklings in a cohort, while 63.8% and 4.9% will be from 5- and 6-egg clutches respectively. Hence, the egg size pattern of 4-egg females may be of adaptive advantage in terms of reproductive success, as the relative duckling size within cohorts determine survival and recruitment [26], and by allowing females in a suboptimal body condition to reduce clutch size and/or abandon ducklings without impairing their survival [cf. 65].

Compliance with general hypotheses

Most studies of intraclutch egg size in eiders and other precocial waterfowl species show variation consistent with the adaptive resource distribution hypothesis. This hypothesis states that females allocate most resources to eggs with a higher probability of survival, which reflects their position in the laying sequence [4,8,9,11,66, see also 36]. In eiders, the highest energy investment, assessed here by larger egg size, is allocated to the first 3 eggs in the laying sequence. Of these, pre-incubation failure (mainly predation) of unattended first laid eggs is much higher than for subsequent eggs [67,68], whereas higher survival in subsequent eggs is attributed to increased nest attendance and incubation initiation, which start after the laying of the second or third eggs [10,67,68]. Hence, increasing the size of the first laid egg when laying 4 eggs, as found in the present study, seems a highly risky investment, if not accompanied by an earlier nest attendance. We did not investigate the timing of nest attendance and incubation initiation in the present study, but Hanssen et al. [10] showed that eiders laying 4-egg clutches start incubation on average one day earlier in the laying sequence than those laying 5-egg clutches. They also showed that an earlier start to incubation was related to poor female body condition, but they did not relate advanced incubation to a larger size of the first laid egg in 4-egg clutches, even though their data showed such a pattern (see Hanssen et al. [10]). In snow geese, Williams et al. [32] reported that larger first laid eggs had a higher survival prior to incubation start than smaller first laid eggs. They did not consider timing of the start of incubation in their study, but their results also indicated that female snow geese laying large first eggs in a clutch were capable of increasing survival of these eggs.

In theory, adaptive changes in egg size when laying a suboptimal clutch size, should occur in eggs of intermediate size and not in the smallest or largest eggs. Hence, in both 4-, 5- (and 6-)egg clutches, expected candidates for adaptive changes in size would be egg number one and three. The present analyses showed a size increase in the relatively large first laid egg, in line with our prediction, but this result somehow conflicts with the resource distribution hypothesis, as first laid eggs generally are reported to have a high failure rate (see above).

Although an earlier incubation start may represent an adaptation to increase first egg success in 4-egg clutches [cf. 10], a size increase in the third laid egg would have been more in line with the resource distribution hypothesis, with largest investment in the two central eggs. However, as previously stated, increasing the size of the third egg would probably postpone incubation initiation and further increase pre-incubation failure (predation risk) of first laid eggs, as incubation initiation may not commence prior to laying the largest egg within a clutch, without inducing a hatching or developmental asynchrony in the embryos [cf. 69]. Alternatively, the laying of two large consecutive first eggs of comparable size may facilitate an earlier incubation initiation, as females may increase nest attendance or start incubation after laying the first of these similar sized eggs, without jeopardizing synchronous development and hatching. If so, the laying of similar sized first and second egg in 4-egg clutches may explain the one-day earlier incubation initiation in these clutch sizes [cf. 10]. Likewise, similar egg size in 6 egg clutches, where the second and third egg are of comparable size, may reflect an adaptation to advance incubation start from the third to the second egg. Although no data exist on the time of incubation start in 6 egg clutches, and sufficient data may be hard to obtain, an advanced incubation start, comparable to that for 5-egg clutches, could explain how reproductive success, expressed as successfully fledged ducklings, generally increases with clutch size [58,59].

In conclusion, our study shows that individual female eiders have a specific within-clutch egg-size strategy, dependent on clutch size. Based on sequential histories of recaptured females, where individual females laid larger first eggs in a 4-egg clutch when in poor condition, and laid smaller first eggs in a 5-egg clutch when in better condition, we argue that this change is a functional adaptation, and that this adaptation may increase reproductive output, as the size increase in first egg results in two ducklings (rather than one) that potentially benefit from large size, in both single or amalgamated broods. As other studies have reported advanced incubation start in females in poor condition laying 4-egg clutches, we hypothesise that the increased investment in first laid eggs, which normally suffer from a high predation rate is, at least partly, compensated by an earlier start of nest attendance and incubation. Hence, our results suggest that eiders possess a finely tuned egg and clutch size strategy, by which individuals can adjust current reproduction in relation to their pre-laying body condition.

Test for the effect of clutch size on whatch.

(DOCX)

Click here for additional data file.

Distribution of clutch sizes.

(DOCX)

Click here for additional data file.

Least square mean estimates (LS means) in mean clutch size for year and test of post hoc pairwise differences.

(DOCX)

Click here for additional data file.

Test of pairwise differences in egg volume between clutch sizes (first to third egg).

Post hoc pairwise tests were estimated with least square mean.

(DOCX)

Click here for additional data file.

27 Feb 2020

PONE-D-19-33049

Condition dependent strategies of egg size variation in the Common Eider Somateria mollissima

PLOS ONE

Dear Mr. Christensen,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Both reviewers appreciated the quality and interest of the manuscript, however, they both raised concerns about the clarity of the analysis  and the structure of the manuscript. Please address their comments carefully.

We would appreciate receiving your revised manuscript by Apr 12 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Julien Martin

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements:

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your Methods section, please provide additional information regarding the permits you obtained for the work. Please ensure you have included the full name of the authority that approved the field site access and, if no permits were required, a brief statement explaining why.

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

4. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This article tests if common eider females adaptively adjust within-clutch egg size depending on clutch size, laying order and their own body condition. The authors use long-term, longitudinal data on a population of common eiders to analyse how variation in clutch size depends on maternal body condition; and how variation in egg size depends on clutch size and laying order, at the population and the individual levels.

The study has potential as the results are of interest, and are based on a strong dataset. The analyses at both the population and the individual levels bring very important information on egg size variation patterns. The study fails, however, to address the hypothesis the authors wish to test: that variation in within-clutch egg size patterns is an adaptive maternal adjustment. The authors analyse the direct relationship between maternal condition and clutch size, and between clutch size, laying sequence and egg size, but do not test the effect of within-clutch egg size variation patterns on any fitness trait (reproductive success in terms of number of hatchlings or offspring survival, or survival of the first egg). They instead rely on literature to make that link. Although literature on birds does state that egg size is related to maternal reproductive success, it has not, from the references listed, been conclusively tested in this population or species, and is not tested in this article. Moreover, laying sequence has been found to be as strong a predictor of hatching success and hatchling survival as egg size, according to some references cited in the manuscript (e.g. ref. 11). While it seems fair to interpret and discuss the results of this study in the larger context of reproductive tactics and adaptive adjustment framework, the aim and hypothesis should be reformulated and nuanced to focus more on and better fit with what the analyses actually test. If data are available to investigate the adaptive value of within-clutch variation (for example by testing survival probability of first eggs in 4- and 5-egg clutches), that should be done.

The methods section needs more work. The data collection part is sufficiently detailed, but the statistics part requires attention. Some of the results presented (clutch size, L202-206 and L216-218) are from analyses that are not described in the methods. Generally, some paragraphs are unclear and need reformulation or more details (e.g. L140-143, or the absence of details concerning linear model selection methods, linear model conformation to assumptions when applicable (homoscedasticity, normality)).

The structure of the manuscript is also at some place surprising, and overall inconsistent. The results are presented in a different order from that in the methods. The same goes for the discussion. The methods section describes an analysis of variation in first, second and third egg sizes across different sizes of clutch (L172-173) in very little details, an analysis of within-clutch egg size variation in relation to laying sequence and clutch size (L174-182), a third analysis of within-individual clutch size variation (L183-191), and a last analysis of within-clutch egg size variation for females with repeated measurements that changed clutch size (L191-194). The results section first presents descriptive results about clutch size and results from a linear model that seem to test the effect of year and female condition on clutch size, but that is not described in the methods section (L201-218). The section then presents descriptive data and results from the analysis body mass and clutch size variation between years for females captured several times (L221-232). Then are presented descriptive data on egg size (L235-239) and results from the linear model testing variation in egg size within clutches, in relation to laying order (L240-271). Finally, the authors present results of within-clutch egg variation according to clutch size for females that changed clutch size in subsequent years (L275-296).

There is also a lack of consistency in terminology (to distinguish tactic from strategy, see Gross 1996, in TREE; egg size and egg volume are used inter-changeably, as are body mass and body condition) and in the way the results are presented (sometimes F values, sometimes effect sizes but referred to as “slope” (L218) or “estimate” (L230), sometimes reporting the R² value of the model (L205)). It would be more informative to present the effect of the variables tested with linear models by giving the effect size (β) and standard error (or any error assessment) rather than F values.

Finally, there are throughout the manuscript several spelling, grammar and punctuation mistakes that need attention.

Detailed comments (with line numbers)

L22-23: the sentence needs to be reformulated: the authors looked at the effect of female body condition on clutch size, and then at the effect of laying sequence and clutch size on egg size, but did not analyse the direct relationship between female body condition and egg size.

L27-28: please reformulate, as “data from recaptured females…” only refers to the individual level of the analyses

L61: please remove the commas: “Alternative but not mutually exclusive hypotheses have related…”

L66: “have proved to fully explain the observed…”

L74-75: what about literature that found a trade-off between size and number of offspring? (even in Anseriformes e.g. Christians 2000 Trade-offs between egg size and number in waterfowl… or Figuerola and Green 2006 A comparative study of egg mass and clutch size in the Anseriformes)

L77-80: sentence unclear, needs reformulation

L86: no need for a comma: “…females is mainly adjusted…”

L86-88: these studies do not show that larger clutches, laid by females in good condition, have a higher reproductive success than small clutches

L89: “records of egg size”

L89-93: are these not results from the analyses described in the paper? It seems so, thus it is surprising to find them presented in the introduction.

L93: please specify the direction of the difference.

L94-96: please provide references

L99: “depending on clutch size”

L100: should be reformulated: the paper analyses the effect of body condition on clutch size, but body condition is not included in the main analysis of the effect of clutch size and laying sequence on egg size.

L108: it seems that all data were collected between 1993 and 2000, so it does not seem relevant to specify “Data on clutch size and egg size”

L109-110: it sounds weird to describe past density in a present tense; this sentence should be reformulated to indicate it was the density estimated at the time the data was collected.

L127: verb missing: “tarsus length […] were recorded”

L135: it is not very clear what “related to the laying sequence” means (numbered? ordered?), this part should be reformulated

L138: the fact that egg size in the MS will refer to egg volume should be introduced here, not L235.

L139 and throughout: please be consistent between body mass and body condition

L139-143: this is unclear, please reformulate

L149-150: it is not clear if body mass refers to actual measured mass or the calculation of body condition

L150: it is unclear if “annual body mass at hatching” refers to maternal or hatchlings body mass

L155-164: this paragraph should be moved to supplementary, and referred to in the paragraph detailing the calculation of female body condition in terms of how the potential effect of such and such variables on body condition was considered and controlled for. Moreover, details are missing concerning the structure of the model ran and the model selection process: it is not clear if the values presented are from a single model. If so, F values of the variables when not involved in the interaction (clutch size and tarsus3) are irrelevant as in the case of an interaction, the effect of the variables involved should not be interpreted separately from the effect of the interaction.

L157: “in which hatching […] was recorded”

L158: please add that Wstart is the response variable for clarity

L165-169: it would be informative to have the number of clutches of each size somewhere (could be in supplementary), to have an idea of the sample sizes of each clutch size

L172-182: this paragraph is not very clear: does the first sentence refer to one analysis (LMM), then L174-182 to another one (ANOVAs)? Do the authors mean they ran repeated measures ANOVAs as post-hoc tests to the linear mixed model? Does individual nest (L178) refer to each egg within a same clutch or an actual nest (and then if a same nest is used several years is it considered as a different nest each year)?

L183: “their reproductive strategy”

L183-191: this paragraph needs to be reformulated for more clarity, it seems that some sentences repeat the same information, and the type of analysis ran should be stated earlier.

L184: it is not clear if a season correspond to a year, or if inter-seasonal changes correspond to yearly changes

L188: it is unclear if individual refer to female ID. Please make sure to distinguish the significance of words that are used for different purpose (individual nest L178)

L191: please be consistent with the terminology used: egg size or egg volume (valid throughout the manuscript)

L205: why present the R² here but not when presenting results from other models?

L213: it is unclear what the authors mean by “shift in dominance”. Is it that in some years there were more 4-egg clutches while in other years 5-egg clutches were dominant?

L218: there are a comma and a space missing between “0.0001” and “slope = 0.006”. Also please be consistent with “,” and “.” when writing numbers.

L227-232: unclear: are females that lost weight more likely to reduce clutch size than females that gained weight, or than when they gain weight? Providing a figure would help represent this result

L241-242: it is unclear if this refers to a difference between first eggs from different clutch sizes, second eggs from different clutch sizes, and third eggs from different clutch sizes, or if they refer to differences between first, second and third eggs across clutch sizes. Does it show the same results as Fig. 2 but in a different way?

L253-257: please state as a bottom note to the table that bold means significant effects and add sample sizes in the title. “data were analyzed with a mixed model”

L261-265: these sentences are a bit stating the obvious… Please reformulate to indicate the direction of the effects.

L267-271: please state as a bottom note to the table that bold means significant effects and add sample sizes in the title. “data were analyzed with a mixed model”

L275-277: please reformulate the sentence to make it clearer

L279-280: please state the direction of the difference

L281-282: please describe the relation documented

L307: “and the second egg marginally smaller than in 5-egg clutches.”

L310-312: the indirect relationship between female body condition and within-clutch egg size variation does not suffice to conclude that egg-size variation patterns have an adaptive value. Plus, laying sequence has been suggested to be as strong a predictor of hatching and nestling survival as egg size. Without testing the effect of such patterns on a measure of fitness, within-clutch egg size variation as a reproductive tactic can at best be interpreted as a hypothetical explanation, hence should be discussed later in the discussion.

L325: is the environment really stochastic?

L325: ‘affecting conditions for breeding females’ is unclear: do the authors refer to the impact of environment on female body condition?

L331-333: please reformulate: the link between egg size and laying order has already been documented, as cited in the introduction (L50-54, L90-91). “for the first time” only applies to the link between within-clutch egg size variation, laying order and clutch size (second part of the sentence)

L343-345: the larger first egg in 4-egg than in 5-egg clutches is significant only in ref 9

L340-345: it is unclear what point this paragraph is making

L352: please remove the parentheses

L363: “predator warning in females”

L364-365: please reformulate, for example: “Such traits likely contribute to an increase in survival”

L366-367: please rephrase this sentence to make it clearer

L371: the formulation “in such a cohort ‘size-scenario’” is unclear

L373-378: are the two percentage distributions statistically different, and statistically different from an expected one based on probabilities? (I’m just curious of the significance of these numbers)

L380: please remove the parentheses

L380-382: that is a really interesting point

L382: “abandon ducklings without impairing their survival”

L387-390: please break the sentence down

L405-406: good point, one potential measure of reproductive success that could be investigated in the present study to test the adaptive value of within-clutch egg size variation, since it was the aim of the study.

L419-415: this paragraph repeats information given in L387-397.

L441-443: well, the study did not test that an increase in first egg size resulted in an increase in the number of large ducklings for 4-egg clutches.

L443-446: this is a hypothesis and not a result of this study, hence this sentence should be more nuanced.

L448-449: there is little information on the link between female body condition and environment in the manuscript, plus common eiders are capital breeders, hence this seems an odd conclusion to the paper.

Fig. 3: presenting these results following the same format as Fig. 2 would make comparisons easier to make for the reader

Reviewer #2: The goal of this paper is to assess whether eiders adjust egg size within the laying sequence depending on clutch size, with the ultimate goal of understanding whether egg size adjustment is adaptive (via optimizing hatching size). This work utilizes an impressive dataset and provides comprehensive evidence that eiders lay first eggs are larger and second eggs smaller in clutches of 4 eggs compared to 5 eggs. I find the paper generally sound and most of my comments are minor and focus on clarifying potential points of confusion.

Line 35 and elsewhere - Rather than wording like “marginally insignificant”, it might be more clear to just say “showed x trend” or “tended to..” just to make the direction of the almost significant relationship more clear.

I found the estimates of female pre-laying body condition quite difficult to follow. Were individual females weighed both at capture and at hatching? Otherwise where does the annual mean mass at hatching come from? Although there are references to mass lost per day of incubation etc, there is no mention as to whether this formula was ad hoc or if there is a precedent for it. It seems to make sense but it takes the reader a while to understand so if possible I’d suggest a bit more detail as to how this particular formula was derived, as well as where the mean mass at hatching comes from.

The wording (particularly in the abstract) about this ‘adaptive’ strategy can come across a bit strong, given the way body condition etc were calculated and without a more comprehensive analysis of female energetics during egg laying/incubation and how it corresponds to increased survival of both the individual female and her chicks. I think the discussion is more moderate in tone so seems appropriate, but I think it would be prudent to adopt that language in the Abstract, rather than stating that this represents a “finely tuned conditional dependent mechanisms that enable females in a suboptimal condition to optimize reproductive output”, which seems a bit over-stated.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Pauline Toni

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

14 Apr 2020

All comments to editor and reviewers are included in the cover letter: Response to reviewers.

Submitted filename: Response to Reviewers2.docx

Click here for additional data file.

8 May 2020

PONE-D-19-33049R1

Condition dependent strategies of egg size variation in the Common Eider Somateria mollissima

PLOS ONE

Dear Mr. Christensen,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by Jun 22 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Julien Martin

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

I found the manuscript much improved and clearer. I sent it back to only 1 of the 2 reviewers who has some more suggestions and minor comments. Please carefully consider them. I probably not send the next revision back to reviewers and accept it if I am happy with the edits.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This manuscript aims at testing how intra-clutch egg size varies with laying sequence depending on clutch size, and makes the link between such variation and maternal body condition, in common eiders. The authors use long-term, longitudinal data to answer these questions at both the population and the individual level. The study is of interest, as the results are sound and based on an impressively large dataset. The analyses at the population and individual levels bring important insights on egg size variation patterns. The paper offers an interesting discussion and interpretation of the results in the larger framework of female reproductive strategies.

This is the second time I review this manuscript (revised submission). I find that the authors have addressed appropriately the concerns that were previously raised, especially the reformulation of the study’s aims, the global organisation and the more modest tone used in the abstract. This revised version is much clearer, more precise, better organised and more readable than the previous one. I only have a few minor comments

The writing could still be improved, notably by shortening sentences, but the discourse is clear enough to be understood. Some English mistakes remain (I attempted at pointing them out in the specific comments).

Regarding the method section, great efforts have been made to provide more details on the analyses and the tests used, and to organise the analyses to mirror the results presentation. I think there is still one analysis description missing: I am guessing the results section entitled “Egg laying patterns within individuals” (L328-353) are from similar analyses to those described in the “intra-clutch egg-size variation” and “patterns of egg size variation in relation to laying order”, but this is not specified in the methods section. Adding one more subsection stating something like “we reproduced the analyses described in sections ‘…’, limiting them to recaptured females that had changed clutch size from 4 to 5 eggs between years” would do the job. Also, I think that, to ensure replicability and transparency, some more details could be provided, for example the models distribution when not using linear models (Poisson?).

As to the statistical features reported, I reckon it is a matter of opinion. I find slopes and SEs more informative as it provides with a direction and a strength of the differences. However, now that the statistics section is more detailed and much clearer than previously, I agree that F-values are inherent to the comparison process used in the analyses of this manuscript. As to the R² values, I have never used SAS, but is it not possible to manually calculate marginal and conditional R² for (G)LMMs, using Nakagawa and Schielzeth (2013 Methods in Ecology and Evolution) approximation? Just so that results are more consistently presented…

Specific comments (line numbers referring to the manuscript version with tracked changes):

Abstract

L22 and throughout: please harmonize between “intraclutch” and “intra-clutch”

L23: please add commas “to clutch size, and the relation between clutch size and female body condition,”

L25-26: I think the authors mean “within the laying sequence depending on clutch sizes in response to body condition”

L26: please reformulate “as such an adjustment could have adaptive implications on”

L28: “advantage for the hatchlings”

L28: “The analyses were first performed at the population level; and then at the individual level using data from”

L31-32: this part is a bit heavy, I suggest reformulating, for example as “(range: 1–6), 4- and 5-egg clutches constituted c.70% of all clutches, and taking turns in being the most represented clutch size.”

L34: “levels”

L39: if I understand correctly, the punctuation should be as follows “pre-laying body condition and clutch size, and the intraclutch”

L40: “pattern indicate that both clutch and egg size are actively”

L43: please nuance by replacing “indicates “with “suggests”

L44: “mechanism” (singular)

L45: “years where females are in suboptimal body condition”

Introduction

L61: “in a population”

L63: “of the first egg”; “efficiency of females”

L67-68: “number, to incubation strategy, to facilitation […] and to differential”

L72: “Although egg size”

L76: “which lay”

L83: by “equal”, do the authors mean “even” (as in ‘make equal’)?

L85: “life history” (no hyphen)

L93-94: “as well as an increase in reproductive success”

L102: “breeding populations, suggesting that eiders could”

L105: “if common eider females change egg size”

L109-110: “data at the population level, as well as the individual level”

Methods

L120: “held” (not holded)

L133-134: “”in cases where no egg-shell remains were found to disqualify the record)

L156: do you mean “hatched eggs” (instead of hatching eggs)?

L171: “To ensure”

L191: is there a difference between general linear model and generalized linear model (L203)? I suspect not, therefore please pick one. Also, if it is a GLM, please specify the response variable error distribution (I am assuming a Poisson one?)

L208: do you mean “Intraclutch” (instead of Interclutch)? Cf. L279

L210: is it a linear mixed model?

L211: “as a fixed effect and year as a random factor”

L215-216: “and year, and interactions between laying order and clutch size, and between year and laying order”

L218: “bypasses”

L220: “egg size could only be”

L221-222: “(e.g. the pattern for the first four eggs could only be tested in clutches”

L222-223: do you mean for each clutch size or that you conducted separate tests on first eggs, then on the first 2 eggs, then the first 3 eggs… (I assume the latter, therefore you might wann reformulate)

L238: “were fulfilled”

Results

L257: “with 4-egg clutches being more frequent in some years”

L261: “Clutch size was”

L263: “0.0001” (instead of 0,0001)

L266 & 268: “The majority of the recaptured females”; “size the following year”

L267 & 269: remove “had” in both instances

L270: “Among the 108 females that did” (no comma)

L272: “and 65 gained”

L274-275: do you mean p < 0.0001 (instead of >)? No capital letter is needed for “slope”

L296: there is one too many closing parenthesis

L301: “data were analysed” (data is plural)

L314: please remove “as indicated by the significant effect of laying order”

L334: with “marked”, do you mean “marginal”?

L338: please replace “evident” with “observed”

Discussion

L364-365: please remove “(although just insignificant)”. I understand you want to remind this point, so I suggest you replace “marginally” by “almost significantly”

L366: “females when they both increased”

L368: please replace “data” with “results”

L378: “laying date varied yearly? by only”

L389: “reflect” (data is plural)

L394: “with” instead of “within”, as I assume you mean females produce eggs according to a specific pattern

L397: “contemporary” is not too clear, I would use “associated pre-laying”

L413: do you mean “in terms of number of ducklings”?

L414: “duckling survival” to what?

L418: “in size with the two”

L422: “the likely associated fitness benefits.”; “waterfowls?” (I am not sure if it is a mistake or if fowl has an irregular plural, same L450)

L433: “Analysed in relation to”

L435: “show” (data is plural)

L441: “5- and 6-egg” (there are some hyphens missing every now and then, also L475, 493, 495)

L442: “in terms of simple size” is not clear, if you mean ‘regarding size’, I think that can be removed as the point is made by the rest of the sentence

L443: please remove “both”, or reformulate as it is unclear which two elements it refers to

L454: either ‘the highest energy investment” or “higher energy investment” (although I think you mean allocation, as you refer to an increase in egg size but not necessarily associated with a greater cost to the female?)

L456-457: “eggs is much higher” (refers to pre-incubation failure, singular) and “eggs is attributed” (refers to survival, singular)

L458: “start” (plural, refers to both nest attendance and incubation initiation)

L502: please replace “specific” with “recaptured”

L503: “when in poor condition”

L504: “we argue that” (no comma)

L508: “clutches, we hypothesise”

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: PAULINE TONI

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

15 Jun 2020

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This manuscript aims at testing how intra-clutch egg size varies with laying sequence depending on clutch size, and makes the link between such variation and maternal body condition, in common eiders. The authors use long-term, longitudinal data to answer these questions at both the population and the individual level. The study is of interest, as the results are sound and based on an impressively large dataset. The analyses at the population and individual levels bring important insights on egg size variation patterns. The paper offers an interesting discussion and interpretation of the results in the larger framework of female reproductive strategies.

This is the second time I review this manuscript (revised submission). I find that the authors have addressed appropriately the concerns that were previously raised, especially the reformulation of the study’s aims, the global organisation and the more modest tone used in the abstract. This revised version is much clearer, more precise, better organised and more readable than the previous one. I only have a few minor comments

The writing could still be improved, notably by shortening sentences, but the discourse is clear enough to be understood. Some English mistakes remain (I attempted at pointing them out in the specific comments).

Regarding the method section, great efforts have been made to provide more details on the analyses and the tests used, and to organise the analyses to mirror the results presentation. I think there is still one analysis description missing: I am guessing the results section entitled “Egg laying patterns within individuals” (L328-353) are from similar analyses to those described in the “intra-clutch egg-size variation” and “patterns of egg size variation in relation to laying order”, but this is not specified in the methods section. Adding one more subsection stating something like “we reproduced the analyses described in sections ‘…’, limiting them to recaptured females that had changed clutch size from 4 to 5 eggs between years” would do the job. Also, I think that, to ensure replicability and transparency, some more details could be provided, for example the models distribution when not using linear models (Poisson?).

The statistics for the section: ‘ Egg-laying patterns within individuals’ was not described in the previous version. When examining it again we realized that our test did not account for the paired design. We therefore reanalyzed the data using a paired t-test. This did not change the significance of the results. We apologize for overlooking this issue earlier.

As to the statistical features reported, I reckon it is a matter of opinion. I find slopes and SEs more informative as it provides with a direction and a strength of the differences. However, now that the statistics section is more detailed and much clearer than previously, I agree that F-values are inherent to the comparison process used in the analyses of this manuscript. As to the R² values, I have never used SAS, but is it not possible to manually calculate marginal and conditional R² for (G)LMMs, using Nakagawa and Schielzeth (2013 Methods in Ecology and Evolution) approximation? Just so that results are more consistently presented…

We do not use R2 estimates for interpreting our results. We doubt that addition of R2 to the generalized linear mixed models would add much information. In addition, the R2 value is not simple to estimate in mixed and generalized mixed models, and the interpretation across statistical models is far from trivial.

Specific comments (line numbers referring to the manuscript version with tracked changes):

Abstract

L22 and throughout: please harmonize between “intraclutch” and “intra-clutch” Corrected

L23: please add commas “to clutch size, and the relation between clutch size and female body condition,” Corrected

L25-26: I think the authors mean “within the laying sequence depending on clutch sizes in response to body condition” Corrected

L26: please reformulate “as such an adjustment could have adaptive implications on” Corrected

L28: “advantage for the hatchlings” Corrected

L28: “The analyses were first performed at the population level; and then at the individual level using data from” Corrected

L31-32: this part is a bit heavy, I suggest reformulating, for example as “(range: 1–6), 4- and 5-egg clutches constituted c.70% of all clutches, and taking turns in being the most represented clutch size.” Corrected

L34: “levels” Corrected

L39: if I understand correctly, the punctuation should be as follows “pre-laying body condition and clutch size, and the intraclutch” Corrected

L40: “pattern indicate that both clutch and egg size are actively” Corrected

L43: please nuance by replacing “indicates “with “suggests” Corrected

L44: “mechanism” (singular) Corrected

L45: “years where females are in suboptimal body condition” Corrected

Introduction

L61: “in a population” Rephrased to “Different eider populations” (as this is shown in several different populations.

L63: “of the first egg”; “efficiency of females” Corrected

L67-68: “number, to incubation strategy, to facilitation […] and to differential” Corrected

L72: “Although egg size” Corrected

L76: “which lay” Corrected

L83: by “equal”, do the authors mean “even” (as in ‘make equal’)? Yes, new phrasing: “make equal”.

L85: “life history” (no hyphen) Corrected

L93-94: “as well as an increase in reproductive success” Corrected

L102: “breeding populations, suggesting that eiders could” Corrected, but the word “could” is not included, as this makes the sentence less precise.

L105: “if common eider females change egg size” Corrected

L109-110: “data at the population level, as well as the individual level” Corrected

Methods

L120: “held” (not holded) Corrected

L133-134: “”in cases where no egg-shell remains were found to disqualify the record) Corrected

L156: do you mean “hatched eggs” (instead of hatching eggs)? Hatching nests refer to nests where ducklings were not hatched, but where eggs were in the process of hatching. Hatched eggs is not a precise description as such nests will be similar to nests where females were captured with ducklings.

L171: “To ensure” Corrected

L191: is there a difference between general linear model and generalized linear model (L203)? I suspect not, therefore please pick one. Also, if it is a GLM, please specify the response variable error distribution (I am assuming a Poisson one?) General linear models assume normal distribution whereas generalized linear models assumes poisson or other non-normal distribution. We have specified the use of non-normal distributions (Generalized linear models) in the statistic paragraph.

L208: do you mean “Intraclutch” (instead of Interclutch)? Cf. L279 As these analyses consider both within and between clutch variation, we have changed the heading to “Inter- and intraclutch egg-size variation in both the Statistical and Result sections.

L210: is it a linear mixed model? As laying order was coded as a categorical variable and as such does not model a linear relation we think that mixed model is the most appropriate term.

L211: “as a fixed effect and year as a random factor” Corrected

L215-216: “and year, and interactions between laying order and clutch size, and between year and laying order” Corrected

L218: “bypasses” Corrected

L220: “egg size could only be” Corrected

L221-222: “(e.g. the pattern for the first four eggs could only be tested in clutches” Corrected

L222-223: do you mean for each clutch size or that you conducted separate tests on first eggs, then on the first 2 eggs, then the first 3 eggs… (I assume the latter, therefore you might wann reformulate) No. There needs to be at least 2 eggs in a clutch for testing laying order. We have rephrased the sentence.

L238: “were fulfilled” Corrected

Results

L257: “with 4-egg clutches being more frequent in some years” Corrected

L261: “Clutch size was” Corrected

L263: “0.0001” (instead of 0,0001) Corrected

L266 & 268: “The majority of the recaptured females”; “size the following year” First part is changed, but the second part (size the following year) is kept, as clutch size change did not in all instances take place from one year to the next. Some birds were not captured every year, but every second year.

L267 & 269: remove “had” in both instances Corrected

L270: “Among the 108 females that did” (no comma) Corrected

L272: “and 65 gained” Corrected

L274-275: do you mean p < 0.0001 (instead of >)? No capital letter is needed for “slope” Yes and Corrected

L296: there is one too many closing parenthesis Corrected

L301: “data were analysed” (data is plural) Corrected

L314: please remove “as indicated by the significant effect of laying order” Corrected

L334: with “marked”, do you mean “marginal”? No. With the t-test applied the significance is marked

L338: please replace “evident” with “observed” Corrected

Discussion

L364-365: please remove “(although just insignificant)”. I understand you want to remind this point, so I suggest you replace “marginally” by “almost significantly” Corrected

L366: “females when they both increased” Corrected

L368: please replace “data” with “results” Corrected

L378: “laying date varied yearly? by only” We have changed “colony” with “annual” to point out that yearly laying dates vary by only 5 days

L389: “reflect” (data is plural) Corrected

L394: “with” instead of “within”, as I assume you mean females produce eggs according to a specific pattern Corrected

L397: “contemporary” is not too clear, I would use “associated pre-laying” Corrected

L413: do you mean “in terms of number of ducklings”? YES- Corrected

L414: “duckling survival” to what? We have specified this: “higher duckling survival is the referred consequences.

L418: “in size with the two” Corrected

L422: “the likely associated fitness benefits.”; “waterfowls?” (I am not sure if it is a mistake or if fowl has an irregular plural, same L450) First part corrected. Second part: new phrasing is “waterfowl species”, which should be the correct (changed both places).

L433: “Analysed in relation to” Corrected

L435: “show” (data is plural) Corrected

L441: “5- and 6-egg” (there are some hyphens missing every now and then, also L475, 493, 495) Corrected

L442: “in terms of simple size” is not clear, if you mean ‘regarding size’, I think that can be removed as the point is made by the rest of the sentence Corrected - deleted

L443: please remove “both”, or reformulate as it is unclear which two elements it refers to Corrected

L454: either ‘the highest energy investment” or “higher energy investment” (although I think you mean allocation, as you refer to an increase in egg size but not necessarily associated with a greater cost to the female?) First part corrected. It should be clear that it is an “allocation” from the previous sentence.

L456-457: “eggs is much higher” (refers to pre-incubation failure, singular) and “eggs is attributed” (refers to survival, singular) Corrected

L458: “start” (plural, refers to both nest attendance and incubation initiation) Corrected

L502: please replace “specific” with “recaptured” Corrected

L503: “when in poor condition” Corrected

L504: “we argue that” (no comma) Corrected

L508: “clutches, we hypothesise” Corrected

Submitted filename: Response to reviewers.docx

Click here for additional data file.

9 Jul 2020

Condition dependent strategies of egg size variation in the Common Eider Somateria mollissima

PONE-D-19-33049R2

Dear Dr. Christensen,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Julien Martin

Academic Editor

PLOS ONE

13 Jul 2020

PONE-D-19-33049R2

Condition dependent strategies of egg size variation in the Common Eider Somateria mollissima

Dear Dr. Christensen:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Julien Martin

Academic Editor

PLOS ONE