Printer Friendly

WITHIN-CLUTCH TRADE-OFFS BETWEEN THE NUMBER AND QUALITY OF EGGS: EXPERIMENTAL MANIPULATIONS IN GULLS.

PAT MONAGHAN [1]

Abstract. We experimentally increased the number of eggs laid by Lesser Black-backed Gulls and examined the relationship between egg number and egg quality. Egg quality was measured in terms of egg composition and the probability that an egg would give rise to a fledged chick. In measuring the latter, we removed the potentially confounding effects (1) between parental quality and egg quality and (2) between increased egg production and parental condition, using a cross-fostering protocol in which eggs were reared by control parents. Lesser Black-backed Gulls were capable of producing, on average, almost three times the normal clutch of three eggs. Although egg mass did not fall below that of the last laid egg in normal clutches, as the clutch extended beyond three, experimental eggs contained relatively less lipid and relatively more water. Overall, the percentage of eggs that gave rise to fledged chicks (when reared singly by a foster parent) declined significantly with position in the experimental laying s equence; about two-thirds of the eggs laid at the beginning of the laying sequence gave rise to fledged chicks, whereas only about one-third of the last laid eggs did so. This was not due to any seasonal change in the capacity of foster parents to rear the chicks, and neither hatching nor fledging success of the eggs was related to their fresh mass. Thus, as more eggs are laid, important changes in egg composition occur that have, in themselves, substantial effects on offspring survival. This study provides clear evidence of a trade-off between egg number and egg quality.

Key words: chick survival; clutch size; egg composition; egg number; egg quality; egg size; Larus fuscus; Lesser Black-backed Gull; life history theory; trade-offs.

INTRODUCTION

Understanding the evolution of the number of young per breeding event is a central issue in life history theory, and identifying the underlying trade-offs remains the focus of a large number of empirical studies across a wide range of taxa (Godfray et al. 1991, Roff 1992, Stearns 1992). One important trade-off influencing the optimal number of offspring is that between propagule quality, often reflected in size, and propagule number. There are sound theoretical grounds for postulating that such a trade-off will exist (Smith and Fretwell 1974). The empirical evidence, however, particularly in animals, is relatively weak, being based largely on correlative studies (reviewed in Roff 1992, Williams 1994, Bernardo 1996). Although such studies can provide important insights into the possible operation of trade-offs, their interpretation requires caution because the potentially confounding effects of, for example, parental quality, have not been controlled. Such confounding effects are likely to be particularly imp ortant in species with post-hatching parental care, such as birds, in which parental quality is known to be a major factor influencing young survival (Amundsen and Stokland 1990, Reid and Boersma 1990, Bolton 1991, Magrath 1992, Blomqvist et al. 1997).

Although there have been numerous experimental manipulations of propagule size and number in plants, comparable manipulations in animals are few, in part due to procedural difficulties (Roff 1992). These have been overcome most notably in studies of lizards (e.g., Sinervo and Licht 1991, Sinervo 1993), in which it has been found that direct manipulation of egg size (either by hormonally altering egg number or by removal of some egg contents) has a profound effect on offspring development and survival. Furthermore, the extent to which egg size actually reflects egg quality is unclear. Changes in size have generally not been separated from changes in nutrient content, and there is a pressing need for empirical data across a range Of taxa on the interrelationships among egg size, egg composition, and offspring fitness (Bernardo 1996).

Surprisingly, despite the fact that much of clutch size theory stems from avian work (Godfray et al. 1991), experimental investigations of the links among egg number, egg quality, and survival of young in birds are few. The fact that direct removal of egg constituents is associated with a very high embryo mortality in birds ([sim]80%; Hill 1993, Finkler et al. 1998), and that the necessary experimental penetration of the avian egg in itself alters embryo metabolism (Finkler et al. 1998), restricts the scope of experimental studies. Another important factor has been the widespread assumption that egg production demands are unlikely to limit clutch size in birds. This has directed the focus of experimental work on reproductive trade-offs toward the chick-rearing phase (Monaghan and Nager 1997). Recent work, however, suggests that egg production in birds can be expensive, in terms of both energy and nutrient requirements, and that experimentally increased egg production has a negative effect on fitness. Formati on of avian eggs requires large amounts of lipids and proteins (Ricklefs 1974, Walsberg 1983, Carey 1996). These demands need to be met at a time of year when food is relatively scarce, and egg production is affected by the amount of energy and quality of nutrients available during this period (e.g., Martin 1987, Arcese and Smith 1988, Boutin 1990, Bolton et al. 1992, Nager and van Noordwijk 1992, Selman and Houston 1996, Ramsay and Houston 1997). The demonstration that experimentally increasing egg production can reduce the capacity of the parents to rear young suggests that the importance of egg production costs in determining optimal clutch size has been underestimated (Heaney and Monaghan 1995, Oppliger et al. 1996, Perrins 1996, Monaghan and Nager 1997, Monaghan et al. 1998). A trade-off between egg quality and egg number in birds therefore seems likely. What is needed is an experimental investigation of the links among egg number, egg quality, and offspring survival, and an experimental design that sepa rates effects on offspring survival mediated through the egg itself from those operating through parental quality.

In some avian species, it is possible to experimentally increase the number of eggs laid to well above the maximum clutch size (Haywood 1993). This procedure can effect changes in egg size and composition (Parsons 1976, Monaghan et al. 1995). It therefore provides the opportunity to carry out the kind of "phenotypic engineering" required to study the trade-off between egg quality and egg number (Sinervo 1993), without the problematic effects hitherto encountered in direct manipulation of avian egg contents. Fostering of the resulting eggs permits one to investigate effects of egg quality on chick survival. Here, we report the results of an experiment based on this technique and designed to investigate the trade-off between egg quality and egg number in the Lesser Black-backed Gull Larus fuscus.

METHODS

This study was carried our in 1996 at South Walney Nature Reserve, Walney Island, northwest England, UK, which supports a large, mixed colony of [sim]24 000 pairs of Lesser Black-backed Gulls Larus fuscus and 8000 pairs of Herring Gulls Larus argentatus. We confined our work to pairs nesting close to the center of the colony, in an area containing [sim]1000 pairs of Lesser Black-backed Gulls. Larus gulls usually lay a clutch of three eggs, but the laying of additional eggs can be experimentally induced by removal of the first and subsequent eggs within 12 h of laying (Parsons 1976, Monaghan et al. 1995); experimental birds then lay extended clutches. To investigate the effects of increasing egg number on egg quality, the experimental eggs were examined in two different ways: (1) their composition and (2) the probability that they would produce a fledgling. We examined both of these indicators of quality in relation to egg size (as indicated by fresh mass). Because we wanted to investigate the trade-off betwe en egg quality and egg number, it was important to ensure that our measure of chick survival was not confounded by any postlaying parental effects operating through a correlation between parental quality or parental condition and egg quality. To remove this, we used a crossfostering procedure. Full details of the experimental protocol will be given.

Manipulation of egg number

At the beginning of the breeding season in early April, all study nests were individually marked prior to egg-laying and were randomly assigned to an experimental or a control group. In the experimental group, we manipulated the number and nutrient content of eggs by continuous egg removal in the following way. Eggs are generally laid at 2-d intervals, and from the start of egg-laying onward, experimental nests were visited at [sim]8-h intervals and any eggs laid were removed. Each removed egg was weighed to the nearest 0.1 g, and maximal width and length were measured. Removal of eggs at a particular nest was continued until that bird stopped laying. In order to ensure that all eggs subsequently laid in a given experimental nest were part of the same clutch, it was necessary to identify the minimum period between the loss of the complete first clutch and the production of a second, new clutch. In a separate sample of birds in which the first clutch was collected after completion, this was found to be 10 d ( range 10-30 d; mean [pm] 1 SE = 12.4 [pm] 1.0 d; n = 19). Accordingly, eggs laid within 10 d of the previous one were considered part of the same clutch. The median interval between two eggs in these continuous laying sequences was 2.7 d (range 1.7-9.3 d; n = 275 eggs) and in only 8.7% of cases was this interval [greater than]4 d. Most pairs (75%) laid all eggs of the extended clutch in the same nest. In the remaining cases, the eggs were laid in a newly built nest on the same territory (usually [less than]0.5 m from the original one). Behavioral observations showed that, in all cases, only one pair was associated with both nests. The clutches collected from the two nests were not larger than clutches in which all eggs had been laid in one nest (t = 0.61, df 30, P [greater than] 0.05). The repeatability of shape (width divided by length) of eggs laid by the same female is typically high (Ojanen et al. 1979, van Noordwijk et al. 1981), and a combination of laying interval and size data is accurate in identifying eggs laid by the same female (McRae 1997). Any sw itching of nests typically occurred about halfway through the extended laying sequence, and the laying interval between the last egg in the original nest and the next egg in the new nest did not differ significantly from the comparable interval in single-nest territories (Mann-Whitney U test: Z = 1.65, P [greater than] 0.05). The shape repeatabilities (coefficients) for extended clutches, in which all eggs came from one nest, and for cases in which the eggs came from two nests on the same territory, were not significantly different (0.420 [pm] 0.008 and 0.411 [pm] 0.042, respectively, mean [pm] 1 SE; t = 0.22, df = 30, P [greater than] 0.05). We generated 10 artificial clutches, each consisting of half of the clutches of two randomly selected females. The egg shape repeatability value (mean [pm] 1 SE) of these composite clutches was 0.109 [pm] 0.009, significantly lower than that of extended clutches laid in two nests on the same territory (t = 7.04, df = 16, P [less than] 0.001). We conclude, therefore, that territory and nest takeovers by other pairs were absent or rare in the study population.

Measurement of egg quality

As previously outlined, egg quality was measured in terms of either egg composition or chick survival. Because these are mutually exclusive procedures, approximately half of the extended clutches were randomly assigned to each method. Clutches that were subjected to each method did not differ in the number of eggs laid (t = 0.23, df = 30, P [greater than] 0.05), laying date (t = 0.71, df = 30, P [greater than] 0.05), mass of the first egg (t = 0.96, df = 30, P [greater than] 0.05), or mass of the last egg (t = 1.05, df = 30, P [greater than] 0.05).

Egg composition analysis.--Chemical analysis of egg composition was performed for all of the eggs of 14 extended clutches (n = 115 eggs in total) and for 14 control clutches of three eggs taken from pairs that initiated egg-laying over a similar period (controls: median initiation clutch date 7 May, range 2 May to 24 May; experimentals: median 8-9 May, range 28 April to 28 May; Mann-Whitney U test: Z = 0.43, P [greater than] 0.05). In order to avoid disrupting the normal laying pattern of the control birds, while ensuring that the eggs were collected shortly after laying, we replaced eggs of the control pairs with wooden dummies at the time of collection of each egg. After being measured, fresh eggs were boiled for [sim]15 mm on the same day, wrapped in cling film and newspaper, and stored in a freezer until further analysis. In the laboratory, eggs were thawed and separated into shell, yolk, and albumen. Each part was separately dried at 60[degrees]C to a constant mass. Because virtually all lipid in the egg is located in the yolk (Ricklefs 1977, Carey 1996), and yolk carbohydrate content is negligible (Ryder et al. 1977, Carey 1996), we determined the lipid content of the yolk only. Lipid was extracted from the yolk using Soxhlet extraction with petroleum ether as a solvent (Dobush et al. 1985). The samples were then dried again and reweighed. Lipid content was determined by subtracting the mass of the dry lipid-free yolk from the dry yolk mass before extraction. In all eggs, thawed egg mass in the laboratory was slightly lower than the fresh egg mass recorded in the field due to some water loss during boiling and freezing. This mass difference was added to the wet albumen mass, as the water loss had come from this component (Williams et al. 1982). The coefficients of variation were similar for all of the egg composition components measured, suggesting no substantial differences in the accuracy with which the different components were measured (10.0% for lean dry mass, 10.7% for dry albumen, 10.9% for dry yolk, 10.6% for lipid, and 9.7% for water; see Table 1).

Hatchability and chick performance.--Egg quality was also assessed in terms of the probability that an egg would give rise to a fledged chick. Because we wanted to examine the quality of the additional eggs independent of any other factors (possible impairment of parental performance due to increased egg-laying demands, competition from other chicks in the brood, differences in parental quality, and seasonal changes in food supply), we adopted the following protocol to remove or minimize these effects. Eggs of 18 experimentally extended clutches (n = 145 eggs in total) were given to foster pairs and allowed to develop naturally. All foster parents were subjected to the same level of disturbance as experimental and control pairs, and foster parents themselves laid a clutch of three eggs. They were chosen such that their laying dates spanned the whole period over which the eggs of the experimental birds were laid (28 April to 7 June). On the day that the foster pairs produced their third egg, this was removed a nd replaced with an egg that had been removed from an experimental nest within the previous 24 h. The other two eggs of the foster pair were prevented from developing by dipping them into mineral oil. Thus, each foster pair laid and incubated a clutch of three, but only hatched the experimental egg; the chick was therefore reared in the absence of any competition from other chicks, and by parents that had not been subjected to any increased egg production.

The probability that the experimental chick would fledge could also be influenced by seasonal differences in the quality of foster parents or in food availability. These factors could affect experimental eggs laid late in the laying sequence, and thereby given to foster parents that had initiated laying relatively late in the season, To minimize possible qualitative differences between the foster parents, our work involved pairs laying at a time when most of the Lesser Black-backed Gulls in our study area were laying clutches of three eggs (77.6% three-egg clutches, n = 804). This implied that we had not yet reached the period in the season, when clutch size declines due to, e.g., parental age effects (Ryder 1980). Only a very weak decline in clutch size was evident in the part of the season over which we were working (quadratic regression, [r.sup.2] = 0.15). We only used as foster parents pairs that had themselves produced a clutch of three eggs, and there was no seasonal tread in fresh egg mass in three-eg g clutches (for the first, second, and third egg, [F.sub.1,514] [less than] 3.50, [r.sup.2] [less than] 0.7%; all P [greater than] 0.05). Because parental quality is reflected in egg mass in this species (Bolton 1991), this further suggests that there was no seasonal decline in the quality of our foster parents. Furthermore, given that the foster parents were required to rear only a single chick, the effect of any remaining variation in parental chick-rearing capacity is likely to have been minimal. To check that this was so, we also cross-fostered the third-laid egg of 43 control nests from one nest to another in the same way, and over the same period (3 May to 6 June), as for experimental eggs. We then examined any seasonal changes in the capacity of the foster parents to rear a single chick produced in a normal clutch laid by another pair.

Around the expected time of hatching, all foster nests were visited daily to check for hatched eggs. The two remaining host eggs that had been prevented from developing were removed when the fostered egg hatched.

On the day of hatching, chicks were individually marked and weighed to the nearest 0.1g. The condition of the chicks on hatching was calculated from a linear regression of body mass on tarsus length, with the observed mass (expressed as a proportion of the predicted mass) taken as the condition index. The survival of the chicks was monitored until day 35, when they were close to fledging. To facilitate the location of chicks, nesting territories were fenced with chicken wire when the chicks were [sim]1 wk old. Hatching success (whether or not the egg hatched) and chick survival (whether or not the chick survived to [geq]35 d) was determined for each foster nest. Embryos and chicks were considered as dead when we found their corpses on the territory. Some eggs (21.1% of 189) and chicks (16.3% of 98) disappeared from the nest and were presumably taken by a predator, most probably another gull. Another four eggs (three experimental and one control egg, 2.1% overall) showed no signs of development, and three egg s (1.6%) were abandoned by their foster parents. These causes of loss showed no seasonal bias and, as they were unlikely to be related to egg quality (which is determined by the original, not the foster, parent), we excluded these cases from our analysis of egg/chick viability.

Instantaneous growth rates during the first week after hatching and during the first three weeks (when growth levels off) were calculated separately for each chick. Growth rates for the first week only were calculated, as it has been shown for Lesser Black-backed Gulls that egg quality can affect chick growth for [geq]1 wk after hatching (Bolton 1991). Instantaneous growth rates R were calculated using the following formula:

R = ([log.sub.10] [W.sub.2] - [log.sub.10] [W.sub.1])/([t.sub.2] - [t.sub.1])

where W is mass and t is time. All fledglings were weighed and their tarsus was measured at day 35. We calculated a condition index for fledglings in the same way as for hatchlings.

Statistical analysis

Throughout the paper, values are expressed as mean [pm] 1 SE. We used two methods to examine the effect of the number of eggs laid on egg quality. First, we used the absolute position of the egg in the laying sequence (i.e., egg 1, 2, 3, etc.). However, egg number 6, for example, will be the last egg in females laying six eggs, but the middle egg in those laying 12. Therefore, we also examined the effect of the relative position of the egg in the laying sequence, calculated as:

relative egg position = (egg number/total number of eggs in clutch) X 100.

The outcome of analyses using absolute or relative position of the egg in the laying sequence did not differ, and we therefore report only the results using absolute position. To analyze within-clutch changes in egg mass or egg composition, we summarized the observed pattern within each extended clutch by using one measure, the slope of the regression on absolute egg number. Analyses were then performed using this summary statistic for each clutch. In order to compare the composition of eggs of different sizes, we used the method suggested by Ricklefs (1984), involving regressions of [log.sub.10] egg component against [log.sub.10] egg mass. We used a logistic regression approach to test for the independent effects of egg mass and laying date (for control eggs) and of egg mass and laying sequence (for experimental eggs) on survival (SPSS, Version 6.02, SPSS, Chicago, Illinois, USA). We tested for nonlinearity by adding the second-order term for egg mass, date, and laying sequence. All interactions between the explanatory variables were tested, and reported when significant. A stepwise backward-elimination procedure was used, and the significance of each variable was tested hierarchically. To test for significance in the logistic regressions, we used changes in deviance. Because the scale parameters (deviance/df) of the final models were always close to 1, the significance of the changes in deviance when removing a factor from the model was tested against the [[chi].sup.2] distribution (Crawley 1992).

RESULTS

Egg mass and composition

As is generally the case in gulls (e.g., Salzer and Larkin 1990, Kilpi et al. 1996), egg size declined with position in the laying sequence in the three-egg control clutches (Fig. la). As detailed in the methods, egg mass in the control clutches did not change with laying date. Within these clutches, absolute values of lean mass, lipid, and water declined over the laying sequence, but their relative contribution as a percentage of fresh egg mass remained constant (Table 1). In addition, within each of first, second, and third eggs, all egg components increased in direct proportion to fresh egg mass ([log.sub.10] : [log.sub.10] regression, t test for slope = 1, all t [less than] 1.16, df = 12, P [greater than] 0.05). On average, an egg from a normal three-egg clutch consisted of 12.7% lean dry mass, 8.1% lipid, and 72.7% water (the remainder being dry egg shell).

The experimental birds laid an average of 8.59 [pm] 0.61 eggs over a period of 23.5 [pm] 1.9 d; the number of eggs produced declined with the date when they first started to lay eggs (r = -0.46, df = 30, P [less than] 0.01). The size (fresh mass) of their first egg (80.7 [pm] 0.75 g, n = 32) did not differ from that of control three-egg clutches (t = 0.28, df 44, P [greater than] 0.05). Within the extended clutches of the experimental females, egg size declined with position in the laying sequence (Fig. 1b). Birds that laid more eggs did not lay a larger first egg (r = -0.03, df = 30, P [greater than] 0.05) but laid a smaller last egg (r = -0.47, df = 30, P [less than] 0.01). The third egg in the extended clutches was larger than a normal third egg (77.5 [pm] 0.91 g, n = 30, and 73.01 [pm] 2.04 g, n = 14, respectively; Mann-Whitney U test, Z 2.36; P [less than] 0.05), whereas the last egg of extended clutches was similar in size to a normal last (i.e., third-laid) egg (75.1 [pm] 0.82 g, n = 32; Fig. la, b; t = 0.43, df = 44, P [greater than] 0.05).

Although egg mass itself did not fall below that in a normal clutch, this was not the case for the nutrients within the eggs. Within-clutch changes in egg composition for the experimental birds are shown in Table 2. Only protein (dry albumen and lipid-free dry yolk) decreased proportionally with decreasing egg mass. Smaller eggs contained disproportionately more water and less lipid; when corrected for egg mass, lipid content declined, water content increased, and lean dry mass (mainly protein) remained constant over the laying sequence. Expressed as percentages of fresh egg mass, egg lipid declined by 0.16 [pm] 0.03% whereas water increased by 0.22 [pm] 0.01% for every additional egg laid. Corrected for differences in egg size, the third egg in extended clutches had the same lipid content as normal third-laid eggs (ANCOVA, [log.sub.10](lipid) on [log.sub.10](mass) for experimental vs. control, with no difference in slopes or elevation and no significant interaction), whereas the last egg contained relativel y less lipid: 6.78 [pm] 0.12% vs. 8.14 [pm] 0.17% in controls (experimentals vs. controls, no significant difference in slope, but significant difference in elevation [F.sub.1,27] = 42.65, P [less than] 0.001). There was no relationship between the lipid content of the first egg and the total number of eggs laid (regression corrected for egg size, t = 0.81, df = 13, P [greater than] 0.05).

Thus, Lesser Black-backed Gulls were able to lay many more eggs than their usual clutch of three without egg size falling below that of normal third-laid eggs. However, as the clutch extended beyond the normal three eggs, experimental eggs contained relatively less lipid (the main energy source for the growing young) and relatively more water.

Foster parent quality

Hatching success in the control cross-fosterings was 69.4% (based on 36 third-laid eggs from normal clutches, having excluded seven losses due to egg predation, which were spread across the season). Hatching success of these fostered eggs did not depend on the egg mass or on the date in the season when the foster parents initiated laying of their own clutch (logistic regression: egg mass, [[chi].sup.2] = 0.003, df = 1, P [greater than] 0.05; laying date, [[chi].sup.2] = 0.84, df = 1, P [greater than] 0.05; interaction not significant). This means that, as expected, there was no seasonal change in the incubation performance of foster parents.

Within this control group, young hatching from larger eggs were heavier and there was a relatively weak tendency for them to be skeletally bigger, independent of the hatching date (multiple regression, n = 22: effect of egg mass on hatchling mass, t = 4.60, P [less than] 0.001; effect of date, t = 0.24, P [greater than] 0.05; effect of egg mass on hatchling tarsus length, t = 1.99, P = 0.06; effect of date, t = 0.65, P [greater than] 0.05). Thus, the condition of hatchlings (mass controlling for skeletal size) increased with increasing egg mass, again independent of hatching date (multiple regression; n = 22; effect of egg mass, t = 3.22, P [less than] 0.01; effect of date, t = 0.85, P [greater than] 0.05). The survival rate until day 35 for chicks from control eggs was 75%; mortalities usually occurred during the first few days (the age when last observed alive was 0.75 [pm] 0.75 d, n = 4). Chick survival was independent of egg mass (logistic regression: [[chi].sup.2] = 0.08, df = 1, P [greater than] 0.05) an d hatching date ([[chi].sup.2] = 0.07, df = 1, P [greater than] 0.05; interaction not significant). Neither growth rate nor fledging mass and condition was related to fresh egg mass (Table 3; [F.sub.1,12] [less than] 0.25, all P [greater than] 0.05). Growth rate, fledging mass, and condition of control chicks also did not change with hatching date ([F.sub.1,12] [less than] 1.33, all P [greater than] 0.05). Thus, there was no seasonal change in the capacity of the foster parents to rear a single chick; hence, any differences in chick survival in the experimental group can be attributed to effects of position in the laying sequence, rather than to laying date.

Survival of experimental chicks

The percentage of experimental eggs that gave rise to fledged chicks declined significantly with position in the laying sequence (Fig. 2). This overall measure of success combines the effects on both hatching and fledging success.

Hatching success was recorded for 95 fertile eggs from the experimentally extended clutches (the three infertile eggs having been excluded; see Methods), of which 74 (77.9%) hatched successfully not significantly different from the 69.4% in the control group; [[chi].sup.2] = 0.60, df = 1, P [greater than] 0.05). As with the control cross-fosterings, hatching success of these eggs did not depend on their size (logistic regression: n = 95; egg mass [[chi].sup.2] = 0.30, df = 1, P [greater than] 0.05). Hatching success did not significantly decline with increasing absolute egg position (Fig. 3a).

As with the control cross-fostered eggs, young that hatched from larger eggs in the experimentally extended clutches were both heavier and skeletally larger (regression analysis; n = 64; effect of egg mass on chick mass, t = 13.73, P [less than] 0.0001; on chick tarsus length, t = 2.38, P = 0.02). There was, however, a significant interaction between the condition of the chick at hatching and the position of the egg in the laying sequence (multiple regression, n = 64; interaction between egg mass and laying position on chick condition, t = 2.54, P = 0.01). Chicks hatching from eggs of a given size but laid later in the sequence were in relatively poor condition.

Despite the effect of egg size on chick condition at hatching, post-hatching chick survival was not related to egg mass. However, chicks hatched from eggs laid earlier in the laying sequence were more likely to survive than chicks that hatched from eggs laid later in the laying sequence, independent of egg mass (Fig. 3b). A chick that hatched from the first egg of an extended laying sequence had an average survival probability of 90.0%, whereas this declined to 31.0% for a chick hatched from the 11th egg.

Growth, fledging mass and fledging condition were not related to egg mass (Table 3; [F.sub.147] [less than] 1.42, all P [greater than] 0.05), and chicks from all eggs reached a similar size at fledging, independent of position in the laying sequence (Table 3). Chick mortality, however, mainly occured early in the nestling period (age when last observed 5.00 [pm] 1.48 d, n = 12). Chicks that survived to fledging initially had a faster instantaneous growth rate (n = 37, median = 70.36, 25th percentile = 60.09, 75th percentile = 75.91) than did chicks that died before leaving the nest (n = 7, median = 49.26, 25th percentile = -9.11, 75th percentile 65.14; unequal variances, Mann-Whitney U test: Z = 2.62, P = 0.01). Mean growth rate during the first week of life also declined with increasing position in the laying sequence (Spearman rank correlation; [r.sub.s] = -0.79, n = 8, P = 0.02; Table 3.) Thus, the negative effects of poor egg quality appear to be centered on the early post-hatching period.

Given that parents were relieved of the incubation and rearing costs in this experiment, it is not very meaningful to examine the relationship between total number of eggs laid and number of offspring fledged. However, the negative effect of laying more eggs on egg quality is highlighted by the fact that there was a negative relationship between the percentage of eggs that gave rise to fledged chicks when reared by foster parents, and the number of eggs that the original parent had laid (Fig. 4). Because all eggs were fostered to different foster parents, this negative relationship must have been due entirely to a decrease in average egg quality as the female continued to lay.

DISCUSSION

The experiment reported here tested the hypothesis that the quality of eggs decreases with increasing egg number. Our experimental procedure extended the number of eggs beyond that in a normal clutch, with experimental birds laying almost three times as many eggs as in their usual three-egg clutch. This increased egg production was not offset by a reduction in egg mass below that of a normal third-laid egg. However, important changes in egg composition occurred as the clutch size extended beyond the normal level, and these changes had negative consequences for chick survival. Egg quality declined progressively, in terms of both nutrient content and the probability of chick survival, as more eggs were laid. Even though all of the rearing costs, with respect to incubation and chick provisioning, were borne by foster parents, and the chicks were reared in the absence of any sibling competition, those females laying more eggs had a reduced return per egg. The return per egg was particularly variable in the large r clutches. There was considerable inter-female variation in egg composition that was also related to the number of eggs laid (the last egg of longer sequences had less lipid), suggesting that egg quality varied more toward the end of long laying sequences. In experimentally increased clutches, we thus found good evidence for a trade-off between egg quality and egg number, and the longer the experimental laying sequence, the poorer the last-laid egg.

There are relatively few data on interactions among egg size, egg composition, and chick survival in birds, and this has been recognized as an important gap in our understanding of the evolution of clutch size (Williams 1994, Bernardo 1996, Erikstad et al. 1998). Larger eggs are generally believed to be more successful in producing a full-grown offspring because they provide the growing young with more and higher quality nutrients (Parsons 1970, J[ddot{a}]rvinen and V[ddot{a}]is[ddot{a}]nen 1983) and confer thermoregulatory advantages to the embryo and the nestling (Rhymer 1988, Wiebe and Bortolotti 1995). In particular, yolk content may influence the quality and viability of hatchlings because it can provide a post-hatching nutritional supplement. Many descriptive studies support this view, but the quality of the parents has frequently been confounded with the effects of egg size (reviewed in Williams 1994). Where the two have been separated, egg size has been found to have some influence on chick survival independently of parental effects, but the effect generally operates only for the first few days after hatching (Nisbet 1978, Bolton 1991, Magrath 1992, Amundsen 1995, Smith et al. 1995, Amundsen et al. 1996). In our study, we used cross-fostering to examine egg effects in isolation from effects operating through the quality of parental care; in addition, any effects due to interactions with other chicks in the brood were also removed, as the chicks were reared singly by their foster parents. The reduced brood size made chick raising easier for the foster parent; this could potentially have offset the effect of egg quality, but it did not. Our test is thus a conservative one with respect to egg quality effects. Egg size, as indicated by fresh egg mass, declined with position in the laying sequence. Chicks from larger eggs hatched in better condition, in line with other studies (Parsons 1970, Bolton et al. 1992, Williams 1994), presumably because they hatched with greater yolk reserves. Nonetheless, egg size d id not influence the probability that a hatched chick would survive to fledging independently of position in the laying sequence. Post-hatching survival, however, was strongly influenced by position in the laying sequence. In agreement with earlier studies (Nisbet 1978, Bolton 1991, Magrath 1992, Amundsen 1995, Smith et al. 1995, Amundsen et al. 1996), egg quality affected the chicks early in the nestling period, through their growth and survival during the first few days. In our experiment, this early disadvantage of chicks hatched from eggs laid late in the laying sequence affected their probability of surviving to fledging, but not the fledging condition of those that did survive. Presumably, as the nestling period progressed, egg quality effects were overridden by the rearing environment, which in this case was provided by foster parents whose egg production costs had not been manipulated and that were only rearing one chick. Overall, around two-thirds of the eggs laid at the beginning of a laying sequenc e resulted in a fledgling, whereas only around one-third of the last eggs reached fledging. Such a decline in return per egg with increasing laying sequence appears to be largely a consequence of poor early growth, and male offspring are particularly vulnerable (Nager et al. 1999).

The effect of position in the laying sequence is likely to have come about through changes in the proportions of different egg components. The extended laying procedure induced changes in the relative nutrient content of the eggs independently of changes in egg size. Our analysis of the composition of eggs showed that, within extended clutches, the relative lipid content declined while the relative water content increased over the laying sequence. Hence, important changes occurred in the nutrient content of the eggs that were not directly reflected in the egg size. This effect has also been found in other studies of the composition of gull eggs, with larger eggs containing disproportionately more albumen and less yolk (Parsons 1976, Ricklefs et al. 1978, Meathrel and Ryder 1987, Meathrel et al. 1987). In our case, variation in albumen content was due to changes in water content and not in protein content. The nutrient contents changed in such a way that a decreased proportion of lipid was "compensated for" t o some extent by an increased water content. Although it has been suggested that water content of the egg may be a more important factor than nutrient content in influencing chick survival (e.g., Finkler et al. 1998), the increased water content in this study did not overcome the effects of changes in nutrient content. Young of eggs laid late in the laying sequence came from eggs with a reduced lipid content, and were thus likely to have been at an energetic disadvantage (Carey 1996). It is possible that other egg components that we did not examine, such as shell structure (Graveland 1990, Roque and Soares 1994, M[ddot{a}]nd 1996, Heaney et al. 1998), or concentrations of substances such as steroid and growth hormones (Schwabl 1997) or immunoglobulins (Apanius 1998) also altered over the laying sequence and contributed to the offspring's lowered survival probabilities. In general, experiments involving manipulation of egg number have not examined the extent to which egg composition was also altered, and the egg size effects reported in such studies must be interpreted cautiously (Bernardo 1996). Our data suggest that, in birds, nutrient content is likely to be more important in determining egg quality than the size of the egg per se.

Egg production is influenced by the female's access to exogenous and endogenous resources (Houston et al. 1995a, b). Deterioration in egg quality over the laying sequence could thus be related to a depressed body condition as the female continues to lay (Alisauskas 1986, Meathrel and Ryder 1987). In gulls, it is believed that the body condition, and in particular the pectoral muscle protein condition, of laying females has an important influence on their egg production capacity (Houston et al. 1983, Hario et al. 1991, Bolton et al. 1993). Females need protein for egg production, including transferring protein to the eggs, maintaining the egg-laying machinery, and probably very importantly, maintaining the condition of the pectoral muscles, which is related to the acquisition of the other resources needed for egg formation. This does not mean that protein content in the egg is what limits chick performance; rather, our study suggests that egg lipid content is very important in chick survival. A decrease in co ndition of the experimental females could explain the observed within-clutch decline in egg quality. It is therefore likely that the number of high-quality eggs that a female bird can lay is limited by physiological processes. In the present study, there was no link between the size or composition of the first-laid (in itself unmanipulated) egg of a particular female and the total number of eggs that she produced. This suggests that females capable of producing large eggs were not necessarily capable of producing more eggs, and that there was no tendency for some females to produce a small number of relatively large, high-quality eggs, or a large number of small eggs.

As an alternative to a proximate, physiological limitation of egg production, adaptive manipulation of nutrient allocation has also been evoked to explain within-clutch variation in egg quality (Williams et al. 1993). According to this hypothesis, there is variation in egg and chick survival with laying sequence, and females allocate more resources to eggs that are more likely to succeed. For gulls that were induced to lay extended clutches, the prospect of raising an offspring successfully might decline with progression through the breeding season or with decreasing parental body condition (which reduces their chick-rearing ability). We found no decline in breeding success that was related to time of season per se (see also Brouwer et al. 1995), but increased egg production can reduce parental performance at later stages of the same breeding attempt (Heaney and Monaghan 1995, Monaghan et al. 1998). Thus, it is possible that, rather than being constrained physiologically, females laying extended clutches may invest fewer resources in the late-laid eggs because they perceive that they have a reduced capacity to rear the chicks successfully. However, this explanation seems unlikely to apply here. Because this study was designed to examine the relationship between egg quality per se and egg production, we excluded any effects of increased production on parental rearing capacity by employing a cross-fostering procedure. We also prevented any sibling competition. Under normal circumstances, the effect of increased egg production on breeding success would be much greater. In addition to poor egg quality, the expected breeding success would further decrease with increased egg production, due to the additional detrimental effects on parental rearing capacity. The laying of only one additional egg has a substantial effect on parental performance (Monaghan et al. 1998). The competitive ability of chicks from poor-quality eggs may also be impaired. It therefore seems unlikely to be worthwhile for females to specifically op t to produce a low-quality egg, which in itself has very poor survival prospects, and which they themselves would have a reduced capacity to rear.

In addition to demonstrating the trade-off between egg quality and egg number, this study also has implications for studies of optimal clutch size. At present, life history theory generally predicts optimal clutch sizes smaller than the most commonly observed clutch size (Roff 1992, Stearns 1992). In this study, we have demonstrated that the production of additional eggs will have negative consequences for the quality of the additional eggs themselves. Brood manipulation experiments, in which, for example, it has been shown that gulls can rear larger broods than their modal clutch size (Harris and Plumb 1965, Haymes and Morris 1977, Winkler 1985), do not account for the effect of laying more eggs on either the eggs or the parents. Failing to consider the costs of egg production, in terms of effects on both egg quality demonstrated here and parental performance demonstrated elsewhere (Heaney and Monaghan 1995, Monaghan et al. 1998), probably leads to an overestimation of the most productive clutch size. Incor porating these negative effects of increased egg production on parental fitness into models of optimal clutch size would lead to lower predictions for optimal clutch size, closer to the observed modal clutch sizes (Monaghan and Nager 1997).

ACKNOWLEDGMENTS

Many thanks to Bill Makin and the Cumbria Wildlife Trust for support at Walney, Andrew Lawson for his help in the field, and Kerry McKay, Pat McLaughlin, and Kenny Ensor for their help in the laboratory analysing egg composition. Thanks also to Neil Metealfe and Graeme Ruxton for constructive criticism of an earlier draft. The work was supported by a grant from the Natural Environment Research Council.

(1.) Address correspondence to this author.

E-mail: P.Monaghan@bio.gla.ac.uk

LITERATURE CITED

Alisauskas, R. T. 1986. Variation in the composition of the eggs and chicks of American Coots. Condor 88:84-90.

Amundsen, T. 1995. Egg size and early nestling growth in the Snow Petrel. Condor 97:345-351.

Amundsen, T., S.-H. Lorentsen, and T. Tveraa. 1996. Effects of egg size and parental quality on early nestling growth: an experiment with the Antarctic petrel. Journal of Animal Ecology 65:545-555.

Amundsen, T., and J. N. Stokland. 1990. Egg size and parental quality influence nestling growth in the shag. Auk 107:410-413.

Apanius, V. 1998. The immune system. Pages 203-222 in J. M. Starck and R. E. Ricklefs, editors. Avian growth and development. Oxford University Press, Oxford, UK.

Arcese, P., and J. N. M. Smith. 1988. Effects of population density and supplemental food on reproduction in song sparrows. Journal of Animal Ecology 57:119-136.

Bernardo, J. 1996. The particular maternal effect of propagule size, especially egg size: patterns, models, quality of evidence and interpretations. American Zoologist 36:216-236.

Blomqvist, D., O. C. Johansson, and F. G[ddot{o}]tmark. 1997. Parental quality and egg size affect chick survival in a precocial bird, the lapwing Vanellus vanellus. Oecologia 110:18-24.

Bolton, M. 1991. Determinants of chick survival in the lesser black-backed gull: relative contribution of eggs ize and parental quality. Journal of Animal Ecology 60:949-960.

Bolton, M., D. C. Houston, and P Monaghan. 1992. Nutritional constraints on egg formation in the lesser black-backed gulls: an experimental study. Journal of Animal Ecology 61:521-532.

Bolton, M., P. Monaghan, and D. C. Houston. 1993. Proximate determination of clutch size in lesser black-hacked gulls: the roles of food supply and body condition. Canadian Journal of Zoology 71:273-279.

Boutin, S. 1990. Food supplementation experiments with terrestrial vertebrates: patterns, problems and the future. Canadian Journal of Zoology 68:203-220.

Brouwer, A., A. L. Spaans, and A. A. N. de Witt. 1995. Survival of Herring Gull Larus argentatus chicks: an experimental analysis of the need for early breeding. Ibis 137:272-278.

Carey, C. 1996. Female reproductive energetics. Pages 324-374 in C. Carey, editor. Avian energetics and nutritional ecology. Chapman and Hall, New York, New York, USA.

Crawley, M. J. 1992. GLIM for ecologists. Blackwell, Oxford, UK.

Dobush, G. R., C. D. Ankney, and D. G. Krementz. 1985. The effect of apparatus, extraction time, and solvent type on lipid extractions of Snow Geese. Canadian Journal of Zoology 63:1917-1920.

Erikstad, K. E., T. Tveraa, and J. O. Bustnes. 1998. Significance of intra-clutch egg-size variation in Common Eider: the role of egg size and quality of ducklings. Journal of Avian Biology 29:3-9.

Finkler, M. S., J. B. van Orman, and P. R. Sotherland. 1998. Experimental manipulation of egg quality in chickens: influence of albumen and yolk on the size and body composition of near-term embryos in a precocial bird. Journal of Comparative Physiology B 168:17-24.

Godfray, H. C. F., L. Partridge, and P. H. Harvey. 1991. Clutch size. Annual Reviews in Ecology 22:409-429.

Graveland, J. 1990. Effects of acid precipitation on reproduction in birds. Experientia 46:962-970.

Hario, M., M. Kilpi, and K. Selin. 1991. Parental investment by the sexes in the Herring Gull: the use of energy reserves during early breeding. Ornis Scandinavica 22:308-312.

Harris, M. P., and W. J. Plumb. 1965. Experiments on the ability of Herring Gulls (Larus argentatus and Lesser Black-backed Gulls (L. fuscus) to raise larger than normal broods, Ibis 107:256-257.

Haymes, G. T., and R. D. Morris. 1977. Brood size manipulation in herring gulls. Canadian Journal of Zoology 55:1762-1766.

Haywood, S. 1993. Sensory and hormonal control of clutch size in birds. Quarterly Review of Biology 68:33-60.

Heaney, V., and P. Monaghan. 1995. A within-clutch trade-off between egg production and rearing in birds. Proceedings of the Royal Society of London B 261:361-365.

Heaney, V., R. G. Nager, and P. Monaghan. 1998. Effect of increased egg production on egg composition in the Common Tern Sterna hirundo. Ibis 140:693-696.

Hill, W. L. 1993. Importance of prenatal nutrition to the development of a precocial chick. Developmental Psycho-biology 26:237-249.

Houston, D. C., D. Donnan, and P. J. Jones. 1995a. Use of labelled methionine to investigate the contribution of muscle proteins to egg production in Zebra Finches. Journal of Comparative Physiology B 165:161-164.

Houston, D. C., D. Donnan, P. J. Jones, I. Hamilton, and D. Osborne. 1995b. Changes in the muscle condition of female Zebra Finches Poephila guttata during egg laying and the role of protein storage in bird skeletal muscle. Ibis 137:322-328.

Houston, D. C., P. J. Jones, and R. M. Sibly. 1983. The effect of female body condition on egg laying in Lesser Black-backed Gulls Larus fuscus. Journal of Zoology 200:509-520.

J[ddot{a}]rvinen, A., and R. A. V[ddot{a}]is[ddot{a}]nen. 1983. Egg size and related reproductive traits in Ficedula hypoleuca breeding in an extreme northern environment. Ornis Scandinavica 14:253-262.

Kilpi, M., L. Hillstr[ddot{o}]m, and K. Lindstr[ddot{o}]m. 1996. Egg size variation and reproductive success in the Herring Gull Larus argentatus: adaptive constrained size of the last egg. Ibis 138:212-217.

Magrath, R. D. 1992. The effect of egg mass on the growth and survival of blackbirds: a field experiment. Journal of Zoology 227:639-653.

M[ddot{a}]nd R. 1996. Increased eggshell porosity in replacement clutches of the Black-beaded Gull Larus ridibundus. Ornis Fennica 73:131-136.

Martin, T. E. 1987. Food as a limit on breeding birds: a life history perspective. Annual Reviews of Ecology and Systematics 18:453-487.

McRae, S. B. 1997. Identifying eggs of conspecific brood parasites in the field: a cautionary note. Ibis 139:701-704.

Meathrel, C. E., and J. P. Ryder. 1987. Intraclutch variation in the size, mass and composition of Ring-billed Gull eggs. Condor 89:364-368.

Meathrel, C. E., J. P. Ryder, and B. M. Termaat. 1987. Size and composition of Herring Gull eggs: relationship to position in laying sequence and the body condition of females. Colonial Waterbirds 10:55-63.

Monaghan, P., M. Bolton, and D. C. Houston. 1995. Egg production constraints and the evolution of avian clutch size. Proceedings of the Royal Society of London B 259:189-191.

Monaghan, P., and R. G. Nager. 1997. Why don't birds lay more eggs? Trends in Ecology and Evolution 12:270-274.

Monaghan, P., R. G. Nager, and D. C. Houston. 1998. The price of eggs: increased egg production reduces offspring rearing capacity. Proceedings of the Royal Society of London B 265:1731-1735.

Nager, R. G., P. Monaghan, R. Griffiths, D. C. Houston, and R. J. Dawson. 1999. Experimental demonstration that offspring sex ratio varies with maternal condition. Proceedings of the National Academy of Sciences (USA) 96:570-573.

Nager, R. G., and A. J. van Noordwijk. 1992. Energetic limitation of the egg laying period of Great Tits. Proceedings of the Royal Society of London B 249:259-263.

Nisbet, I. C. T. 1978. Dependence of fledging success on egg size, parental performance and egg composition among common and roseate terns Sterna hirundo and S. dougallii, Ibis 120:207-215.

Ojanen, M., M. Orell, and R. A. Vaisanen. 1979. Role of heredity in egg size variation in the Great Tit Parus major and the Pied Flycatcher Ficedula hypoleuca. Ornis Scandinavica 10:22-28.

Oppliger, A., P. Christe, and H. Richner. 1996. Clutch size and malaria resistance. Nature 381:-565.

Parsons, J. 1970. Relationship between egg size and post-hatching chick mortality in the Herring Gull (Larus argentatus). Nature 228:1221-1222.

Parsons, J. 1976. Factors determining the number and size of eggs laid by the Herring Gull. Condor 78:481-492.

Perrins, C. M. 1996. Eggs, egg formation and the timing of breeding. Ibis 138:2-15.

Ramsay, S. L., and D. C. Houston. 1997. Nutritional constraints on egg production in the blue tit: a supplementary feeding study. Journal of Animal Ecology 66:649-657.

Reid, W. V., and P. D. Boersma. 1990. Parental quality and selection on egg size in the Magellanic Penguin. Evolution 44:1780-1786.

Rhymer, J. M. 1988. The effect of egg size variability on thermoregulation in mallard (Anas platyrhynchos) offspring and its implication for survival. Oecologia 75:20-24.

Ricklefs, R. E. 1974. Energetics of reproduction in birds. Pages 152-292 in R. A. Paynter, editor. Avian energetics. Publication Number 15 of the Nuttall Ornithological Club, Cambridge, Massachusetts, USA.

Ricklefs, R. E. 1977. Composition of eggs of several bird species. Auk 94:350-356.

Ricklefs, R. E. 1984. Variation in the size and composition of eggs of the European Starling. Condor 86:1-6.

Ricklefs. R. E., D. Hahn, and W. A. Montevecchi. 1978. The relationship between egg size and chick size in the Laughing Gull and Japanese Quail. Auk 95:135-144.

Roff, D. A. 1992. The evolution of life histories. Chapman and Hall, New York, New York, USA.

Roque, L., and M. C. Soares. 1994. Effects of egg shell quality and broiler breeder age on hatchability. Poultry Science 73:1838-1845.

Ryder, J. P. 1980. The influence of age on the breeding biology of colonially nesting seabirds. Pages 153-168 in B. L. Olla and H. E. Winn, editors. Behavior of marine animals. Plenum Press, New York, New York, USA.

Ryder, J. P., D. E. Orr, and G. H. Saedi. 1977. Egg quality in relation to nest location in Ring-billed Gulls. Wilson Bulletin 89:473-475.

Salzer, D. W., and G. J. Larkin. 1990. Impact of courtship feeding on clutch size and third-egg size in glaucous-winged gulls. Animal Behaviour 39:1149-1162.

Schwabl, H. 1997. Maternal steroid hormones in the egg. Pages 3-13 in S. Harvey and R. J. Etches, editors. Perspectives in Avian Endocrinology. Journal of Endocrinology, Bristol, UK.

Selman, R. G., and D. C. Houston. 1996. The effect of pre-breeding diet on reproductive output in zebra finches. Proceedings of the Royal Society of London Series B 263:1585-1588.

Sinervo, B. 1993. The effect of offspring size on physiology and life history. BioScience 43:210-218.

Sinervo, B., and P. Licht. 1991. Hormonal and physiological control of clutch size, egg size, and egg shape in side-blotched lizards (Uta stansburiana): constraints on the evolution of lizard life histories. Journal of Experimental Zoology 257:252-264.

Smith, C. C., and S. D. Fretwell. 1974. The optimal balance between size and number of offspring. American Naturalist 108:499-506.

Smith, H. G., T. Ohlsson, and K. J. Wettermark. 1995. Adaptive significance of egg size in the European Starling: experimental tests. Ecology 76:1-7.

Stearns, S. C. 1992. The evolution of life histories. Oxford University Press, Oxford, UK.

van Noordwijk, A. J., L. C. P. Keizer, J. H. van Balen, and W. Scharloo. 1981. Genetic variation in egg dimensions in natural populations of the great tit. Genetica 55:221-232.

Walsberg, G. E. 1983. Avian ecological energetics. Pages 161-220 in D. S. Farner, J. R. King, and C. S. Parkes, editors. Avian Biology. Volume 7. Academic Press, New York, New York, USA.

Wiebe, K. L., and G. R. Bortolotti. 1995. Egg size and clutch size in the reproductive investment of American Kestrel. Journal of Zoology 237:285-301.

Williams, A. J., W. R. Siegfried, and J. Cooper. 1982. Egg composition and hatching precocity in seabirds. Ibis 124:456-470.

Williams, T. D. 1994. Intraspecific variation in egg size and egg composition in birds: effects on offspring fitness. Biological Review 68:36-59.

Williams, T. D., D. B. Lank, and F Cooke. 1993. Is intraclutch egg-size variation adaptive in the Lesser Snow Goose? 0ikos 67:250-256.

Winkler, D. W. 1985. Factors determining a clutch size reduction in California Gulls (Larus californicus): a multi-hypothesis approach. Evolution 39:667-677.
           Composition of eggs from 14 control clutches of three
               eggs laid by Lesser Black-backed Gulls (mean
              [pm] 1 SE) in relation to the laying sequence.
Egg                  Egg sequence in clutch
component [+]                 First                  Second
Dry yolk mass (g)    [10.30.sup.a] [pm] 0.22 [10.10.sup.a] [pm] 0.28
  Percentage                 12.85 [pm] 0.16         13.00 [pm] 0.19
Dry albumen mass (g)  [6.29.sup.a] [pm] 0.16  [6.04.sup.a] [pm] 0.14
  Percentage                  7.84 [pm] 0.12          7.79 [pm] 0.14
Lean dry mass (g)    [10.30.sup.a] [pm] 0.21  [9.80.sup.b] [pm] 0.21
  Percentage                 12.83 [pm] 0.08         12.64 [pm] 0.16
Lipid mass (g)        [6.30.sup.a] [pm] 0.16  [6.39.sup.a] [pm] 0.20
  Percentage                  7.85 [pm] 0.13          8.17 [pm] 0.13
Water mass (g)       [58.52.sup.a] [pm] 1.24 [56.24.sup.b] [pm] 1.11
  Percentage                 72.91 [pm] 0.24         72.50 [pm] 0.21
Egg                                               F
component [+]                 Third          (df = 2, 12)
Dry yolk mass (g)     [9.26.sup.b] [pm] 0.27    17.79 [***]
  Percentage                 12.66 [pm] 0.22     2.70
Dry albumen mass (g)  [5.90.sup.b] [pm] 0.22     4.16 [*]
  Percentage                  8.04 [pm] 0.16     1.18
Lean dry mass (g)     [9.24.sup.c] [pm] 0.31    28.65 [***]
  Percentage                 12.60 [pm] 0.14     1.10
Lipid mass (g)        [5.97.sup.b] [pm] 0.18     8.99 [**]
  Percentage                  8.14 [pm] 0.13     2.60
Water mass (g)       [53.48.sup.c] [pm] 1.54    17.57 [***]
  Percentage                 73.00 [pm] 0.19     2.28


Notes: Differences between eggs within clutches were tested using repeated-measures ANOVA ((*.)P [less than] 0.05; (**.)P [less than] 0.01; (***.)P [less than] 0.001). Mean values followed by different superscript letters were significantly different at the level of 5% (except for dry albumen at 7%).

(+.)Percentage values (percentage of wet mass) were arcsine-transformed prior to analysis.
               Summary of a multiple regression analysis of
                within-clutch changes in egg composition in
                experimentally extended clutches of Lesser
                            Black-backed Gulls.
              Slope estimate                   Slope estimate
Egg            for egg mass                    for egg number
component          Mean       1 SE   t [+]          Mean       1 SE
Dry yolk          0.4384     0.1286 4.37 [***]    -0.0071     0.0001
Dry albumen       1.2455     0.1729 1.42           0.0028     0.0011
Lean dry mass     0.9956     0.0700 0.06           0.0006     0.0009
Lipid             0.3863     0.1661 3.70 [**]     -0.0143     0.0056
Water             1.0849     0.0212 4.00 [**]      0.0010     0.0001
Egg
component      t [++]
Dry yolk      7.54 [***]
Dry albumen   2.66 [*]
Lean dry mass 0.68
Lipid         2.55 [*]
Water         7.19 [***]


Notes: For each clutch (n = 13, one clutch with too few eggs for testing), we calculated the slope of the [log.sub.10] : [log.sub.10] regression of the egg component on egg mass and on egg number. These were tested for differences from a slope of 1 in the case of egg mass (egg component increases proportionally with increasing egg mass) and from a slope of 0 in the case of egg number (egg component corrected for egg mass changes over the laying sequence).

(*.)P [less than] 0.05;

(**.)P [less than] 0.01;

(***.)P [less than] 0.001.

(+.)[H.sub.0]: mean slope = 1.

(++.)[H.sub.0]: mean slope = 0.
                Instantaneous growth rate (mean [pm] 1 SE)
                   during the first week after hatching
                    (initial) and until mass levels off
                  (overall), fledging mass, and fledging
               condition of control chicks and experimental
                  chicks in relation to their position in
                           the laying sequence.
                  Laying
                 sequence     Initial growth [+]      Overall growth [+]
                   (egg   ([log.sub.10] [mass]/d) ([log.sub.10] [mass]/d)
Treatment        number)       X [10.sup.3]            X [10.sup.3]
Control                       65.28 [pm] 3.11         54.79 [pm] 2.28
Experimental        1         70.13 [pm] 3.89         51.83 [pm] 3.02
                    2         68.18 [pm] 4.02         55.53 [pm] 2.75
                    3         67.59 [pm] 3.49         48.01 [pm] 6.60
                    4         71.08 [pm] 4.88         53.56 [pm] 2.05
                    5         59.86 [pm] 4.98         57.62 [pm] 2.23
                    6         37.38 [pm] 27.49        29.84 [pm] 23.10
                    7         56.18 [pm] 3.59         52.96 [pm] 10.17
                    8 [+]     52.85 [pm] 35.24        43.92 [pm] 13.44
Correlation [ss]
 n                                  8                       8
 [r.sub.s]                         -0.79                   -0.33
 P                                  0.02            [greater than]0.05
                      Fledging           Fledging
Treatment             mass (g)          condition [++]
Control             636 [pm] 35.4    0.963 [pm] 0.055
Experimental      753.5 [pm] 34.7    1.041 [pm] 0.034
                  708.2 [pm] 56.4    0.997 [pm] 0.037
                  726.3 [pm] 101.2   0.939 [pm] 0.103
                  724.6 [pm] 87.0    1.007 [pm] 0.069
                  705.0 [pm] 34.2    1.020 [pm] 0.052
                  891.3 [pm] 93.8    1.114 [pm] 0.038
                       865.0              0.945
                  662.0 [pm] 42.3    0.979 [pm] 0.061
Correlation [ss]
 n                       8                8
 [r.sub.s]              -0.10            -0.19
 P               [greater than]0.05 [greater than]0.05


Notes: See Methods for calculations. Effect of the position in the laying sequence on growth and size of experimental chicks was tested using regression analyses.

(+.)Note that values have been multiplied by [10.sup.3].

(++.)Fledgling condition was calculated from a linear regression of body mass on tarsus length, with the observed mass (expressed as a proportion of the predicted mass) taken as the condition index.

(ss.)Spearman rank correlation of the mean value and position in the laying sequence.

The relationship between the percentage of eggs giving rise to fledged chicks (hatching plus fledging success combined) when reared singly by foster parents, and the position of the egg in the experimentally extended laying sequences of Lesser Black-backed Gulls ([r.sub.s] = -0.74, n=8, P = 0.037). Samples sizes of eggs at each position arc given within the histograms.
Laying sequence (egg number) Success (%)
1                                14
2                                12
3                                 9
4                                 9
5                                 8
6                                10
7                                 6
8+                               11
COPYRIGHT 2000 Ecological Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

 
Article Details
Printer friendly Cite/link Email Feedback
Author:NAGER, RUEDI G.; MONAGHAN, PAT; HOUSTON, DAVID C.
Publication:Ecology
Geographic Code:1USA
Date:May 1, 2000
Words:10383
Previous Article:ERRATA.
Next Article:SURVIVAL RATES OF A NEOTROPICAL PARROT: IMPLICATIONS FOR LATITUDINAL COMPARISONS OF AVIAN DEMOGRAPHY.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters