The timing of egg laying, clutch size, and egg size of the American Oystercatcher (Haematopus palliatus) were studied over six consecutive breeding seasons in Virginia. Synchrony of laying dates occurred in each of five localities of the study area in at least one year. Mean clutch size was 2.8 eggs (mode = 3) in first clutches and 2.4 eggs (mode = 2) in replacement clutches. Individual females laid replacement clutches of the same size and laid eggs of similar average volume in all years. A change in mate had little effect on the date on which females initiated their first clutches in successive years. The average egg size in a clutch was correlated with the size of the laying female. Egg-size ordering occurred within clutches, the first-laid egg being smaller than the second egg and about equal in volume to the third. We propose that the second egg is largest because it has the highest probability of hatching, and the resulting sibling hierarchy reduces the frequency of sibling competition. Received 19 October 1983, accepted 19 April 1984.

Department of Zoology, University of Toronto, Toronto, Ontario M5S 1A1, Canada; and 2Department of Ornithology, Royal Ontario Museum, 100 Queen's Park Crescent, Toronto, Ontario M5S 2C6, Canada SYNCHRONY of clutch-initiation dates, uni- form egg size, and four-egg clutches are as- sumed to be adaptations of many shorebirds (Charadrii and Scolopaci) for living within the constraints of high-latitude breeding seasons. Clutch-initiation dates occur over a very brief period (Holmes 1971), and replacement clutch- es are uncommon (Pitelka et al. 1974), because the period of abundant food resources is brief and the season is telescoped (Holmes 1972, Nettleship 1973, Pitelka et al. 1974). A defini- tive clutch of four eggs of similar size and shape apparently forms the optimal configuration for minimizing the rate of heat loss when the clutch is uncovered (Norton 1970). Uniform egg size within a clutch also implies no differential allocation of parental care to the chicks (Miller 1979). Uniform chick size may be as important a consequence of four similarly sized eggs as that provided by the energetic advantage during incubation, particularly be- cause a definitive clutch of four is also found in many shorebirds nesting in the temperate zone (Maclean 1972). Oystercatchers (Haematopodidae) are one of the few families of shorebirds whose young de- pend almost exclusively on the adults for food. 3 Present address: 364 Waterloo Avenue, Guelph, Ontario NIM 3K2, Canada. 855 Clutch size in this group is usually less than four eggs, and this may be related to the exten- sive parental care (Maclean 1972). Egg-size or- dering within a clutch is undocumented in oystercatchers, but it might be expected to oc- cur in them rather than in shorebirds that do not feed their young. This is because, through preferential feeding, oystercatcher parents have a greater potential for "control" over which of their chicks survive (Alexander 1974). This po- tential control might be extended to the eggs, particularly if food is limiting or variable (Howe 1976), if preferential feeding is impractical, and if the probability of raising all chicks is depen- dent on the food supply. The purpose of this paper is to document variations in clutch size and egg size in Amer- ican Oystercatchers (Haematopus palliatus) and to elucidate seasonal and annual patterns in the timing of egg laying. By comparing individ- ually known females over several breeding seasons, we have attempted to gauge repeat- ability measures (Falconer 1981) of these char- acteristics, as well as the effect of female size. METHODS Study area.--Study sites were located on Wallops and Assawoman islands, Virginia (37ø50'N, 75ø35'W) in 1978-1983 and on the salt marshes and dunes around the Chincoteague Channel, Virginia (37ø55'N, 75ø23'W) in 1981-1983. For purposes of comparing WALLOPS STATION MAIN BASE I S L A'"'N D / øø' ATLANTIC OCEAN KM o 1 2 3 I I I I Fig. 1. Map of study area including five breeding sites: (1) Chincoteague Island (mean of 5 nests per year), (2) Chincoteague Point (œ = 13 nests), (3) northern Wallops Island (œ = 7 nests), (4) the sand spit of southern Wallops Island (œ = 8 nests), and (5) Assawoman Island (œ = 6 nests). the synchrony of clutch-initiation dates among lo- calities, we divided the study area into five localities: (1) Chincoteague Island, (2) Chincoteague Point, (3) northern Wallops Island, (4) the sand spit of south- ern Wallops Island, and (5) Assawoman Island (Fig. 1). Fieldwork was conducted at the study sites within the period ! March to 31 July for six successive breeding seasons from 1.978 to 1983. Observations in all seasons began by at least 6 April. Data collection.--Over the study period, 60 adults (30 females, 30 males) were trapped at the nests with drop traps lined with fish net (Mills and Ryder 1979). The adults were banded with aluminum bands and unique color-band combinations and were weighed and measured. From 1978 to 1980, all territories were searched periodically, although we were not always certain of clutch sequence (i.e. whether first or re- placement clutch) or the order of egg laying. To ob- viate this problem, we searched territories through- out the prelaying and laying periods in the 1981- 1983 breeding seasons. Eggs were marked, weighed, and measured. All weights used in the analyses are of eggs weighed within 48 h of being laid. Weight can be difficult to determine in the field, particularly on windy days, whereas length and breadth measures are more pre- cise. We therefore wanted a method for predicting volumes. Sixty-eight eggs from museum collections were filled with water to determine their volumes. Regression coefficients were calculated with volume as the dependent variable and the product of egg length (EL) times breadth 2 (EB 2) as the independent variable (Vaisanen 1969). The resultant formula was used to calculate the volume of all eggs measured: Volume (cm 3)=0.47736 x EL (mm) x EB 2 (mm )- 1.318(mm3); r 2 = 0.96. Fresh egg weight (EW) was a good predictor of the calculated volume: Volume (cm 3) = 0.757 EW (g) + 5.96 (cm3); r  = 0.87, n = 170. In the subsequent text, egg size and volume are used synonymously. Statistical analysis.--The clearest method for detect- ing the effects of the female, clutch sequence, and year on egg size is to look at the effects of each factor in two two-way designs (Sokal and Rohlf 1981). We tested for the effects by using a mixed-model, two- way analysis of variance (ANOVA) with females as a random factor, and assumed no interaction be- tween females and the fixed factor (e.g. year or clutch sequence). We then estimated the variation attribut- able to females and to the fixed factor and the sig- nificance levels of the two factors by using the inter- action term as the error term. We used mean values to avoid the confounding effects of different cell sizes produced by differently sized clutches. When using year as the main effect, we examined only first clutches to avoid any confounding of clutch se- quence in the design. In the comparison of means, we used t-tests on every pair of means, while con- trolling the experiment-wise error rate by lowering the value of a to a/n where n is the number of paired comparisons made. To test for differences in egg size across years, across clutches within a year, and within differently sized clutches we used a one-way analysis of variance. We also tested for differences in egg size as a result of egg order by means of a one-way analysis of vari- The ratio of the between-individual component of variance to the total variance measures the correla- tion between repeated measurements of the same in- dividual and is known as the repeatability of a char- acter (Falconer 1981). Heritability, the resemblance between relatives, may be much less than repeat- ability, but it cannot be greater (Falconer 1981). The within-individual component arises from environ- mental fluctuations between successive measures and the systematic effects of age. The between-individual component comprises both permanent environmen- tal effects and a genetic component (Findlay and Cooke 1982). We calculated repeatability measures by using the results of a one-way analysis of variance applied to (1) egg size among females across years, (2) egg size among females across clutches within a year, and (3) date of first clutch initiation across years. A three-way contingency table was constructed to observe the pattern of egg-size ordering in succes- sive years. The table was analyzed using log-linear models after the example of Fienberg (1970) and Bishop et al. (1975). In addition to testing the as- sumption of mutual independence, which character- izes classical contingency analyses, the techniques of Fienberg can be used to calculate the frequencies of events that would be expected under all possible conditions of dependence (interaction) between two or more classifying variables. Goodness-of-fit tests are then used to determine which of a series of hierar- chical models that are based upon various assump- tions of dependence best explain the observed fre- quencies of the event (Harder 1980). If social stimulation tends to promote synchrony in the initiation of first clutches in neighbors, then females in a locality should tend to lay at more sim- ilar dates than females in different localities. One way to test for synchronization is to compare the variance of observed clutch-initiation dates of birds in one area (Sl 2) with the variance of observed clutch- initiation dates of all other birds (s2 ) and to compute the ratio of variances (s2/s?). We obtained 500 ran- dom partitions of the data into samples of size n (the number of nests in the locality in question) and n2 (the number of total clutches initiated in a year - n), computed the variance ratio for each, and then looked to see if our observed value fell in the upper 5% tail of the distribution of 500 empirically obtained ratios (Sokal and Rohlf 1981). RESULTS Clutch initiation.--American Oystercatchers breeding in Virginia initiated clutches over a short period. Initiation dates of first clutches ranged from as early as 6 April (1981) to as late as 13 May (1982); the mean spread of dates for 1981-1983 was only 25 days, however. The av- erage date of initiation of first clutches for 1981- 1983 was progressively later in the successive breeding seasons (Fig. 2). Frequency distribu- tions of clutch-initiation dates before 28 April are similarly shaped in the 1981 and 1982 .-I o I.l.I m 14 I0 6 2 14 I0 6 2 14 I0 6 2 I00 50 0 M  1983 M 1982 1981 '"*' .... *"'"'..... 2 eggs 3eggs 14 22 30 8 16 24 I 9 17 APRIL MAY JUNE Fig. 2. Frequency histograms for 1981-1983 breeding seasons illustrating number of first clutches (open bar), second clutches (gray bar), third clutches (crosshatched bar), and clutches of unknown sequence (bold stippled bar). Location of mean (Z) and median (M) laying dates are indicated by arrows. The bottom panel shows the proportion of two- and three-egg clutches (n = 156). breeding seasons. In 1983, the curve is flat- tened from an increase in the spread of clutch- initiation dates, and the mean date of first clutch initiation is slightly later. The 1983 field season was characterized by a cold, wet April, and this most likely caused the shift in clutch-initiation dates. The time of initiation of replacement clutch- es was related to the date on which the first clutch was lost. High spring tides during 24- 26 April 1981 caused most nests to be lost, and a peak of renesting occurred from 12 to 18 May. In 1982, the spring tides occurred before the majority of females had laid and after eggs in the first clutch had hatched. In 1983, nests were lost to both spring tides and predation at about the same time. This pattern of nest loss resulted in a small second peak of laying from 5 to 12 May. The pairs represented by this peak lost their nests a second time to predation, all with- TABLE 1. Clutch size of Haematopus palliatus on Chincoteague, Wallops, and Assawoman islands, Virginia in six breeding seasons, 1978-1983. Number of eggs Number of eggs in Number of eggs in Number of eggs in in clutches of first clutch second clutch third clutch unknown number Year n ! 2 3 4 u a 1 2 3 u a 1 2 3 u a 1 2 3 u a 1978 33 0 1 2 1 0 0 0 0 0 0 0 2 9 18 1979 43 1 14 18 0 1 3 4 0 1 1 0 0 0 1980 18 0 0 0 0 0 0 0 0 0 0 2 6 10 1981 83 1 2 31 6 1 17 11 5 0 4 1 3 0 1 0 1982 54 0 4 31 6 0 9 1 2 0 1 0 0 0 0 1983 63 0 4 19 13 1 8 6 1 0 6 1 1 2 0 1 Total 294 2 25 101 1 25 3 37 22 8 0 12 3 3 5 18 28 1 clutches with unknown number of eggs. in 4 days, resulting in a third peak of laying in late May and early June (Fig. 2). Approximately three quarters of the total variance in clutch-initiation dates was ex- plained by the effects of female and year. About 59% of the variation in clutch-initiation dates was explained by variation between females (P < 0.001) and about 16% by variation be- tween years (P < 0.001). The repeatability (ra) measure for date of clutch initiation for those females that laid in three consecutive years was high (total phenotypic variance = 97.8, ra = 0.73, n = 22) and indicates that females tend to lay at about the same time each year. An exami- nation of the year-to-year correlations between clutch-initiation dates, however, makes clear the facts that in 1983 females were less likely to lay at the same dates as in 1981 and 1982 (1981- 1982, r = 0.740, P < 0.0001, n = 22; 1982-1983, r = 0.485, P < 0.02, n = 23) and were also less likely to lay on the same dates relative to their neighbors (Spearman rank correlation coeffi- cients: 1981-1982, r = 0.816, P < 0.0001; 1982- 1983, r = 0.463, P < 0.02). Again, a cold wet April and persistent high tides in some nesting areas caused some females to lay at a later date in 1983 than in previous years, and this fact alone would lower the year-to-year correlation. When known females on a territory attained new mates (n = 4) in successive breeding sea- sons, they still tended to lay at about the same date as in the previous year. In the two in- stances when males retained their territories but acquired different mates in the subsequent breeding season, however, these new females laid 16 and 26 days later than the original res- idents. Synchrony of clutch-initiation dates of first clutches (as defined by a significant s22/s 2 ratio, see methods) occurred in all five localities in at least one of the three years of intensive study (1981-1983). Females nesting on the sand spit of southern Wallops Island in 1982 were more synchronous than the general population (six nests, P < 0.05). There was evidence of syn- chrony in all three years among the females on Chincoteague Point (nine nests: 1981, 0.05 < P < 0.1; 1982, P < 0.01; 1983, P < 0.01). At the north end of Wallops Island, there was significant synchrony in 1983 (four nests, P < 0.01) but not in either 1981 or 1982. Synchrony was also detected in 1983 on Chincoteague Is- land (5 nests, P < 0.01) and in 1981 on Assa- woman Island (5 nests, P < 0.01). Clutch size.--The average clutch size declined as the season progressed. Most first clutches in 1981-1983 contained three eggs (Table 1). About two-thirds of first replacement clutches (sec- ond clutch) contained two eggs. Second re- placement clutches (third clutch) contained two eggs in all but three nests. In 1979, 42.4% of first clutches contained two eggs, and in 1978 one clutch (3.0% of all clutches) contained four eggs. The proportion of three-egg clutches de- clined through the breeding season (Fig. 3), but females laid three-egg clutches as late as 1 June in 1981 and 2 and 13 June in 1983 (Fig. 2). Fe- males normally lay a maximum of two replace- ment clutches when preceding clutches are lost, but one female laid three replacement clutches in 1979, although one and possibly two of these were incomplete. Females that laid replace- ment clutches in 1981 and 1983 tended to lay TABLE 2. Egg dimensions (mean + SE) of first, second, and third laid eggs of H. palliatus in Virginia. Egg sequence  Eggs of 1 2 3 unknown number (n = 116) (n = 95) (n = 69) (n = 241) Volume (cc) 41.68 + 0.27 a 43.22 + 0.27 b 42.28 + 0.32 a 42.34 + 0.18 Length (ram) 56.05 _+ 0.21 a 57.09 + 0.19 b 56.95 + 0.20 b 56.31 + 0.13 Breadth (ram) 39.46 +_ 0.11 a 39.82 + 0.11 a 39.43 + 0.14  39.68 + 0.07 Weight (g) 47.69 + 0.42 a 49.65 _+ 0.40 b 47.72 q- 0.53 a 49.30 + 0.50  Homogeneous groups indicated with superscripts. Differences tested using Multiple t-tests (P < 0.015) following one-way ANOVA (Sokal and Rohlf 1981). the same sized clutches (X 2 = 3.19, df = 1, n = 13, 0.05 < P < 0.10). Egg s/ze.--Egg size in the combined data set (including data for all eggs of known laying sequence, regardless of clutch number or clutch size) varied systematically with laying order (Table 2). On average, the second egg laid was the largest (most voluminous) and the heaviest. The first and third eggs were about equal in volume. When we categorized the data accord- ing to clutch sequence, we obtained a similar result. In first clutches (a sample of mostly three-egg clutches), the second egg was larger and heavier than the first (P < 0.015) and heavier than the third (P < 0.015). The first and third eggs were of equal size, but the third egg was longer (P < 0.015). In second clutches (a sample of two-egg and three-egg clutches), the second egg was also the largest (P < 0.015). We then categorized the data by the size of the clutch. In three-egg clutches, the second egg was largest and heaviest and the first and third eggs were equal in size and weight (P < 0.015). The second egg was longer than the first (P < 0.004) and wider than the third (P < 0.011). In two-egg clutches, the second egg was largest (P < 0.02) and tended to be longer than the first (P < 0.07). The pattern of egg-size ordering varied by year (Table 3). In 1982, the second egg was the same size as the first about half the time, and there was no significant difference in their vol- umes. The log-linear model (Fienberg 1970) that best fits the overall pattern of egg-size ordering by year indicates an interactive effect between year (Y) and egg-size order (E) and an inde- pendent effect of clutch size (C) (Table 4). The standardized deviates, which indicate the de- viation from the value expected based on the TABLE 3. Frequencies and standardized deviates a of clutches with first eggs greater than and equal to or less than second eggs in two- and three-egg clutches from 1978 to 1983. Egg order Year Clutch size 1 >_ 2 1 < 2 Total 1978- Two-egg 1 (-0.1) 3 (-0.2) 4 1979 Three-egg 3 (0.8) 5 (-0.3) 8 1981 Two-egg 2 (-1.1) 18 (1.3) 20 Three-egg 3 (-1.7) 25 (0.5) 28 1982 Two-egg 4 (0.4) 4 (-1.9) 8 Three-egg 12 (2.6) 16 (-0.3) 28 1983 Two-egg 2 (-0.1) 10 (1.3) 12 Three-egg 3 (-0.4) 9 (-0.7) 12 Total 30 90 120  Standardized deviates = (Obs. Exp)/EX/- for model of complete independence (Bishop et al. 1975). TABLE 4. Values of the log-likelihood ratios (G 2) for selected models of interaction between clutch size (C), presence of egg-size ordering (E), and year (Y) in the data from Table 3. For a given model, vari- ables that appear together within commas were as- sumed to be jointly dependent, but the effect of that interaction is independent of the other spec- ified interactions or single factor effects? Model G 2 df P E, C, Y 19.99 10 0.05 EY 15.05 8 0.05 < P < 0.1 C, EY 6.41 7 n.s. E, CY 14.16 7 0.05 Y, EC 19.21 9 0.05 EC, EY 5.63 6 n.s. EC, EY, CY 0.53 3 n.s.  Probabilities less than 0.05 that are associated with models of equal or less complexity indicate their in- ability to provide adequate explanations of the data (Sokal and Rohlf 1981). TABLE 5. Effect of clutch size on variation in egg-size parameters (mean ñ SE). Clutch size 2-egg 3-egg Egg (n = 68) (n = 286) pa Volume (cc) 42.4 + 0.29 42.8 ñ 0.15 Length (mm) 56.9 + 0.25 56.6 + 0.12 Breadth (mm) 39.5 + 0.14 39.8 + 0.05 Weight (g) 48.7 + 0.44  48.7 -+ 0.27 0.1 0.05) to explain the data. The proportion of clutches with no apparent egg-size ordering was considerably higher in 1982 than in any other year. Average egg volume did not differ between clutches of two and three eggs (Table 5). In a comparison of egg size in first and second clutches laid by the same female, volume, breadth, and weight were significantly lower in second clutches than in first, and most of the total variation was attributable to differences among females (Table 6). In a similar compar- ison among all clutches of known sequence, however, no differences in egg size between first and second clutches emerged, but length and volume were significantly lower in third clutches than in first or second clutches (first clutch: length = 56.8 + 1.61 mm, volume = 42.7 + 2.19 cc; second clutch: length = 56.4 + 1.86 mm, volume = 42.2 + 2.37 cc; third clutch: length = 55.1 + 1.43 mm, volume = 40.6 + 2.14 cc, one-way ANOVA, P < 0.015 for differ- ences as above). Mean egg volume for each female was plot- TABLE 6. Effect of female, clutch sequence, and year on variation in egg size parameters. A. Clutch size (n = 39 females) Variation attributable to First Second Female a Clutch sequence a clutch clutch Egg (œ) (œ) R  (%) F R  (%) F Volume (cc) 43.8 42.5 71.6 3.69*** 9.0 17.67'** Length (mm) 56.7 56.3 76.1 3.38*** 0.1 2.40 Breadth (mm) 40.2 39.8 74.5 4.05*** 7.1 14.74'** Weight (g) 50.0 48.2 78.5 4.39* * * 2.8 4.79* B. Year (n = 24 females) Variation attributable to Female' Year' Egg 1981 1982 1983 R = F R = F Volume (cc) 43.0 42.7 42.8 55.2 2.48* * 0.38 0.20 Length (mm) 56.7 56.6 56.7 84.6 11.04'** 0.07 0.09 Breadth (mm) 39.8 39.7 39.7 60.8 3.13'* 0.35 0.21 Weight (g) 48.4 48.6 50.3 55.1 2.18' 8.80 4.00 a Significance levels designated as follows: *** = P < 0.001, ** = P < 0.01, * = P < 0.05. Differences among means tested by two-way ANOVA of females with clutch sequence and year. TABLE 7. Analysis of phenotypic variation in average egg size over several breeding seasons and over successive clutches in each breeding season. Volume Length Breadth Egg size over several breeding seasons (n = 24 females, 72 observations) Between-individual variance 1,560.77 (added variance component due to differences among 99's) Within-individual variance 2,942.32 Total phenotypic variance 4,503.09 Repeatability (ra) 0.35 Egg size over successive clutches 1981 (n = 23 females, 48 observations) Between-individual variance 3,410.46 Within-individual variance 3,132.05 Total phenotypic variance 6,542.51 ra 0.52 1983 (n = 15 females, 37 observations) Between-individual variance 1,116.23 Within-individual variance 2,788.20 Total phenotypic variance 3,904.43 ra 0.29 2.28 0.326 0.59 0.427 2.87 0.753 0.80 0.43 1.89 0.43 0.95 0.54 2.84 0.97 0.66 0.44 1.64 0.47 1.70 0.28 3.34 0.75 0.49 0.63 ted with the date of clutch initiation, and no apparent relationship was found (1981, r = 0.13, n = 25, P > 0.05; 1982, r = -0.16, n = 23, P > 0.05) except in 1983, when volume was nega- tively correlated with date (r = -0.39, n = 29, P < 0.006). This result probably reflects the larger number of second replacement clutches of lower average volume in 1983. About 55% of the variation in egg size over the three years, 1981-1983, was attributable to differences among females rather than differ- ences among years (Table 6). There was no in- dication of a difference in egg size in any of the six years of the study (P > 0.05, one-way ANOVA with full data set). Females for which we have data for 3-6 years appear to lay eggs of about the same average volume; there is no apparent change as a result of increasing age. Repeatability estimates of egg characteristics over several years were available for 24 females (Table 7A). About 35% of the total phenotypic variance in egg volume in first clutches arises from differences between individuals. The re- mainder reflects intraindividual variation in response to fluctuating environmental condi- tions. Within a single season, between 29 and 66% of the total phenotypic variance arises from differences between females (Table 7B), de- pending on the year and the egg character. Female size.--Females were heavier than males (females, œ = 638 + 42 g, n = 30; males, œ=567 _+ 113g, n = 30; t = 2.75, P < 0.01). Av- erage egg volume (EV) ranged from 35.85 cc to 45.18 cc (œ = 41.94 + 0.504 SE). To test for a re- lationship between female size and egg vol- ume (EV), we first used body weight as an in- dex of size. Body weight and EV were correlated significantly (r = 0.47, P < 0.05, n = 19). Body weights of oystercatchers are known to fluctu- ate markedly in the breeding season, however, especially around the egg-laying period (Mer- cer 1968). We therefore also tested for a corre- lation between EV and the geometric mean (GM) of eight external body measurements, (Mosimann 1970, Mosimann and James 1979), this geometric mean being free of short-term seasonal bias. Female size as represented by GM is correlated significantly with EV (r = 0.508, P < 0.025) but accounts for only 4% more of the variation in egg volume than does weight. Whether one uses body weight or the GM of body measurements, it is apparent that larger females tend to lay more voluminous eggs than do smaller ones, although clearly female size was not the only factor affecting egg size in our study. Similar analyses of male measurements and weights revealed no influences of male size on egg size. TABLE 8. Comparison of nest-initiation dates in 11 species of shorebirds. Mean num- ber Range of oh- (num- serva- her of Species tions days) Location Comments Reference American Oystercatcher 28 25 37ø50'N, 75ø30'W ( Haematopus palliatus ) Black Oystercatcher 8 21 48ø38'N, 123ø17'W (H. bachmani) 17 24 48ø18'N, 123ø32'W African Black Oystercatcher 55 84 33ø05'S, 17ø57'E (H. moquini) European Oystercatcher 45 42 51ø42'N, 5ø16'W Mean of 3 (H. ostralegus) yr Black-bellied Plover 3 5.5 75ø40'N, 84ø35'W Mean of 4 ( Pluvialis squatarola) yr Willet 80 21 37ø40'N, 75ø20'W Est. (by (Catoptrophorus semipalmatus) Not), 2 yr  Ruddy Turnstone 13 13 81ø49'N, 71ø18'W 1 yr ( A renaria interpres ) Western Sandpiper 58 26 61ø30'N, 64ø50'W Mean of 3 ( Calidris mauri) yr  White-rumped Sandpiper 6 12.5 68ø40'N, 101ø59'W Mean of 2 (C. fuscicollis) yr Dunlin 12 9 61ø31'N, 64ø50'W Mean of 3 (C. alpina) yr 18 22 61ø30'N, 21ø40'E Mean of 5 yr Stilt Sandpiper 14 14 58ø45'N, 93ø00'W Mean of 3 ( C. himantopus) yr b Mean of 3 yr Total of 5 yr I yr This study Drent et aL (1964) Hockey (1983) Keighley and Bux- ton (1948) Hussell and Page (1976) Howe (1982) Nettleship (1973) Holmes (1972) Parmelee et at. (1968) Holmes (1971) Soikketi (1967) Jeht (1973) Replacement clutches not identified. Range estimated from hatching dates. DISCUSSION Most first clutches in our study area were initiated within a 20-day period in all inten- sively studied breeding seasons (1981-1983). The oystercatchers we studied were slightly less synchronous in the spread of initiation dates for first clutches than a population of breeding Willets ( Catoptrophorus semipalmatus) found nearby (Howe 1982, Table 8) but more syn- chronous than H. moquini in South Africa and H. ostralegus in Scotland (Table 8). As expected, shorebirds nesting at higher latitudes are more synchronous in clutch-initiation dates than are American Oystercatchers (Table 8), although interspecific differences in nesting synchrony cannot be accounted for solely on the basis of latitude differences (e.g.H. ostralegus vs. H. pal- liatus). Social stimulation probably affects laying dates in oystercatchers. Birds within localities had lower variances in the dates of initiation of their first clutches than did the general pop- ulation in the study area. Synchrony within lo- calities was most consistent where the largest numbers of oystercatchers nested (Chinco- teague Point). Oystercatchers are very vocal and aggressive toward neighboring pairs, frequent- ly participating in long piping displays in the prelaying period. This extensive piping may stimulate egg laying in some females. Consis- tent differences in environmental factors af- fecting females in different localities probably do not explain this synchrony, as neighboring pairs of oystercatchers used widely differing foraging sites, and these differences were as great as the differences between sites (Not in prep.). The high repeatability estimate ob- tained here is confounded by the effect of so- cial stimulation, and no inference about a her- itable component can be made. Most replacement clutches contained two eggs. With some exceptions, most clutches ini- tiated after 3 June contained two eggs. The av- erage volume of eggs in later clutches was smaller, and this is probably a result of food limitations later in the season. Similarly, in Red- billed Gulls (Larus novaehollandiae) in New Zea- land, egg volume and clutch size declined as the season progressed, and, in that case, the decline appeared to be closely related to the decline in the amount of food (swarms of planktonic euphausiids) over the breeding sea- son (Mills 1979). The food of American Oyster- catchers does not fluctuate as rapidly as do zoo- planktonic populations, but possibly it is reduced enough through oystercatcher preda- tion (O'Connor and Brown 1977) to affect both the size and number of eggs in replacement clutches. Females tend to lay the same sized replace- ment clutch from year to year. This consistency in clutch size over several years probably re- suits from consistent differences in territory quality among females rather than from a ge- netic component. Birds in poor territories may be contributing a greater proportion of their total energy reserves to a 3-egg first clutch and then can lay only a 2-egg replacement clutch, whereas birds occupying richer feeding terri- tories can lay a second 3-egg clutch. This is cur- rently under study. Over a third of the yearly and annual varia- tion in egg size arose from the between-indi- vidual component of the total phenotypic vari- ation. At least part of this component is attributable to the effect of female size. A high repeatability in egg size from year to year has been found in at least six species of shorebirds (Viisinen 1969, Viisinen et al. 1972, Miller 1979), in Darwin's finches (Grant 1982), and in Great Tits (Parus major; Ojanen et al. 1979, Van Noordwijk et al. 1981). In at least some species, egg size has also been found to be heritable [Great Tit, Van Noodwijk et al. 1981; Red Grouse (Lagopus lagopus scoticus), Moss and Watson 1982]. Within-clutch ordering of egg size in shore- birds has received considerable attention, pri- marily because eggs laid last in other Charad- riiformes (notably larids) are usually smaller, and this pattern may have considerable adap- tive significance (Coutson 1963, Parsons 1970, Gochfeld 1977, Lundberg and Viisinen 1979). The literature on shorebird egg size is con- fused, however, because of the apparent equa- tion of increased size with increased length or breadth. In American Oystercatchers, an in- crease in length does not necessarily influence volume and, therefore, the size of an egg. In those studies in which researchers found an increase in only one of the components of vol- ume with laying sequence (e.g. Miller 1979, Not and Lambert 1984), egg size as measured by volume has usually remained constant (e.g. Cairns 1977). Thus, a special explanation for the adaptivehess of egg-size ordering in most shorebirds with four-egg clutches need not be sought. In American Oystercatchers, the first egg is smaller than the second egg but about equal to the third egg in three-egg clutches and smaller than the second egg in two-egg clutches. Eggs hatch in the order that they are laid. There are at least three hypotheses to explain the adap- tive significance of egg-size ordering, but most explain only consistent increases or decreases in egg size over the laying sequence (e.g. larger later eggs being laid because of increased pre- dation of incomplete clutches, Warham 1974, Miller 1979, Clark and Wilson 1981; higher provisioning in later laid eggs enhancing the competitive position of chicks hatched from those eggs in the case of hatching asynchrony, Howe 1976; smaller, later eggs being laid as a form of brood reduction, Parsons 1975, Goch- fetd 1977, Braun and Hunt 1983). The explanation for the larger second egg in clutches of the American Oystercatcher may re- side with the different risks to each of the eggs. Incubation is commenced after the second egg is laid (pers. obs.), and thus for the initial 2 days the first-laid egg is exposed to increased risk of loss from predators and environmental fluctuations. The probability of loss of the first egg is likely to be higher than that of the eggs in an incubated clutch (Miller 1979). Although eggs hatch in the order they are laid, the first two eggs hatch relatively synchronously, and the resultant young may leave the nest before the third egg has hatched. Disturbance from predators or other sources can cause the par- ents to abandon the third egg to attend the hatched young. The second egg, therefore, may have the highest probability of survival, and this could account for the pattern of egg-size ordering within clutches. The magnitude of egg-size differences (4% by volume) in American Oystercatchers is small compared with those in most larids (10-30% of volume, see Parsons 1970, Davis 1975, Nisbet 1978, Lundberg and Viisinen 1979) but similar to that recorded in Roseate Terns (Sterna dou- gallii) late in the breeding season (Nisbet and Cohen 1975). In American Oystercatchers, fresh egg weight and the weight of the young at hatching are correlated (r = 0.82, n = 8, P < 0.025), the heaviest chick coming from the sec- ond egg. Within broods, the differences in chick weights parallel those among egg weights, un- like the situation with Black-legged Kittiwakes (Rissa tridactyla), in which asynchronous hatch- ing markedly enhances weight differences within the brood once all chicks have hatched (Braun and Hunt 1983). Even relatively small differences in egg size can result in a sibling social hierarchy. The sib- ling social hierarchy in H. palliatus is similar to that described previously for H. ostralegus and H. bachmani (Groves 1978, Safriel 1981). When a parent arrives with food, one chick rushes toward that parent while the others remain crouched or standing with the second parent. If the other chicks run to be fed, the dominant chick turns and chases its siblings. Only when the dominant chick seems to be satiated does the next in rank emerge to be fed. In H. ostral- egus, the social hierarchy among chicks in a brood follows the weight hierarchy at hatching (Safriel 1981), and it is probable that this is true of H. palliatus too. The social hierarchy facilitates brood reduc- tion during periods of low food availability, because subordinate chicks either starve or are eliminated by predators (Safriel 1981). In pe- riods of food abundance, all chicks in the brood can be fed. The sibling hierarchy also might promote a more efficient distribution of food if less of the parents' and chicks' energy is wasted during sibling competition (Hamilton 1964, Hahn 1981). This advantage alone might pro- vide an impetus for parental manipulation of egg size independent of the predictability of resources. ACKNOWLEDGMENTS This study was supported by Natural Sciences and Engineering Research Council grants to A. J. Baker. We thank C. Risley, E. Cadman, S. Nash, J. Mc- Donnell, T. Clark, J. Koehler, A. Rivers, B. Howard, and D. Richardson for field assistance and C. and E. Risley for kindly providing accommodation and hos- pitality. Access to Wallops Island was granted by Na- tional Aeronautics and Space Administration officials at the Wallops Island facility and is gratefully ac- knowledged. G. Watson kindly allowed volumetric determinations of the eggs at the Smithsonian Insti- tute, and A. Zimmerman loaned us her boat. The manuscript benefitted from appraisals by J. Barlow, G. Bortolloti, R. Viisinen, A. Van Noordwijk, and an anonymous referee. We also thank S. Baker for draw- ing the figures and M. Goldsmith for typing the manuscript. LITERATURE CITED ALEXANDER, R.D. 1974. The evolution of social be- havior. Ann. Rev. Ecol. Syst. 5: 325-383. BISHOP, Y. g. g., $. E. FIENBERG, & P. W. HOLLAND. 1975. Discrete multivariate analysis. Theory and practice. Cambridge, Massachusetts, MIT Press. BRAUN, B. M., & G. L. HUNT, JR. 1983. Brood reduc- tion in Black-legged Kittiwakes. Auk 100: 469- 476. CAIRNS, W. E. 1977. Breeding biology and behavior of the Piping Plover (Charadrius melodus) in southern Nova Scotia. Unpublished M.Sc. thesis. Halifax, Nova Scotia, Dalhousie Univ. CLARK, A. B., & D. $. WILSON. 1981. Avian breeding adaptations: hatching asynchrony, brood reduc- tion and nest failure. Quart. Rev. Biol. 56: 253- 277. COULSON, J.C. 1963. Egg size and shape in the kit- tiwake (Rissa tridactyla) and their use in estimat- ing age composition of populations. Proc. Zool. Soc. London 140: 211-227. DAVIS, J. W. F. 1975. Age, egg-size and breeding success in the Herring Gull Larus argentatus. Ibis 117: 460-473. DRENT, R., G. F. VAN TETS, E. TOMPA, & K. VERMEER. 1964. The breeding birds of Mandarte Island, British Columbia. Can. Field-Natur. 78: 208-263. FALCONER, D. $. 1981. Introduction to quantitative genetics, Second ed. London, Longman. FIENBERG, $. E. 1970. The analysis of multidimen- sional contingency tables. Ecology 51: 419-433. FINDLAY, C. $., & F. COOKE. 1982. Breeding syn- chrony in the Lesser Snow Goose (Anser caeru- lescens caerulescens). I. Genetic and environmen- tal components of hatch date variability and their effects on hatch synchrony. Evolution 36: 342- 351. GOCHFELD, g. 1977. Intraclutch egg variation: the uniqueness of the Common Tern's third egg. Bird-Banding 48: 325-332. GRANT, P.g. 1982. Variation in the size and shape of Darwin's finch eggs. Auk 99: 15-23. GROVES, S. 1978. Sibling rivalry and its effect on growth of Black Oystercatcher chicks. Pacific Seabird Group Bull. 5(2): 69. HAHN, D.C. 1981. Asynchronous hatching in the Laughing Gull: cutting losses and reducing ri- valry. Anim. Behav. 29: 421-427. HAMILTON, W. D. 1964. The genetical evolution of social behaviour: I, II. J. Theor. Biol. 7: 1-52. HARDER, L.D. 1980. Winter use of montane forests by porcupines in southwestern Alberta: prefer- ences, density effects, and temporal changes. Can. J. Zool. 58: 13-19. HOCKEY, P. A. R. 1983. Aspects of the breeding bi- ology of the African Black Oystercatcher. Os- trich 54: 26-35. HOLMES, R. T. 1971. Latitudinal differences in the breeding and molt schedules of Alaskan Red- backed Sandpipers (Calidris alpina). Condor 73: 93-99. 1972. Ecological factors influencing the breeding season schedule of Western Sandpipers (Calidris mauri) in subarctic Alaska. Amer. Midi. Natur. 87: 472-491. HOWE, H. F. 1976. Egg size, hatching asynchrony, sex, and brood reduction in the Common Grack- le. Ecology 57: 1195-1207. HOWE, M.A. 1982. Social organization in a nesting population of eastern Willets (Catoptrophorus semipalmatus). Auk 99: 88-102. HUSSELL, D. J. T., & G. W. PAGE. 1976. Observations on the breeding biology of Black-bellied Plovers on Devon Island, N.W.T., Canada. Wilson Bull. 88: 632-653. JEHL, J. R., JR. 1973. Breeding biology and system- atic relationships of the Stilt Sandpiper. Wilson Bull. 85: 115-147. KEIGHLEY, J., & E. J. M. BUXTON. 1948. The incuba- tion period of the Oyster-catcher. Brit. Birds 41: 261-266. LUNDBERG, C., & g. A. V(IS.NEN. 1979. Selective correlation of egg size with chick mortality in the Black-headed Gull (Larus ridibundus). Condor 81: 146-156. MACLEAN, G.L. 1972. Clutch size and evolution in the Charadrii. Auk 89: 299-324. MERCER, g. J. 1968. Individual weight change in breeding oystercatchers. Bird Study 15: 93-98. MILLER, E.H. 1979. Egg size in the Least Sandpiper Calidris minutilla on Sable Island, Nova Scotia, Canada. Ornis Scandinavica 10: 10-16. MILLS, J. A. 1979. Factors affecting the egg size of Red-billed Gulls Larus novaehollandiae scopulinus. Ibis 121: 53-67. ß & J.P. RYDER. 1979. Trap for capturing shore and seabirds. Bird-Banding 50: 121-123. MOSIMANN, J.E. 1970. Size allometry: size and shape variables with characterizations of the lognor- mal and generalized gamma distributions. J. Amer. Stat. Assoc. 65: 930-945. --, & F. C. James. 1979. New statistical methods for allometry with applications to Florida Red- winged Blackbirds. Evolution 33: 444-459. Moss, R., & A. WATSON. 1982. Heritability of egg size, hatch weight, body weight, and viability in Red Grouse (Lagopus lagopus scoticus). Auk 99: 683- 686. NETTLESHIP, D. N. 1973. Breeding ecology of turn- stones Arenaria interpres at Hazen Camp, Elles- mere Island, N.W.T. Ibis 115: 202-217. NISBET, I. C. T. 1978. Dependence of fledMing suc- cess on egg-size, parental performance and egg- composition among Common and Roseate terns, Sterna hitundo and S. dougallii. Ibis 120: 207-215. --, & M. E. COHEN. 1975. Asynchronous hatch- ing in Common and Roseate terns, Sterna hitundo and S. dougallii. Ibis 117: 374-379. NOL, E., & A. LAMBERT. 1984. Comparison of Kill- deers breeding in mainland and peninsular sites in southern Ontario. Can. Field-Natur. 98: 7-11. VAN NOORDWlJK, A. J., L. C. P. KEIZER, J. H. MAN BAL- EN, & W. SCHARLOO. 1981. Genetic variation in egg dimensions in natural populations of the Great Tit. Genetica 55: 221-232. NORTON, D.W. 1970. Thermal regime of nests and bioenergetics of chick growth in the Dunlin (Calidris alpina) at Barrow, Alaska. Unpublished M.S. thesis. College, Alaska, Univ. Alaska. O'CONNOR, R. J., & R. A. BROWN. 1977. Prey deple- tion and foraging strategy in the Oystercatcher Haematopus ostralegus. Oecologia 27: 75-92. OJANEN, M., M. ORELL, & R. A. V(ISNEN. 1979. Role of heredity in egg size variation in the Great Tit Parus major and the Pied Flycatcher Ficedula hy- poleuca. Ornis Scandinavica 10: 22-28. PARMELEE, D. F., D. E. GREINER, & W. D. GRAUL. 1968. Summer schedule and breeding biology of the White-rumped Sandpiper in the central Cana- dian Arctic. Wilson Bull. 80: 5-29. PARSONS, J. 1970. Relationship between egg size and post-hatching chick mortality in the Herring Gull (Larus argentatus). Nature 228: 1221-1222. 1975. Asynchronous hatching and chick mortality in the Herring Gull Larus argentatus. Ibis 117: 517-520. PITELKA, F. A., R. T. HOLMES, & S. F. MACLEAN, JR. 1974. Ecology and evolution of social organi- zation in Arctic sandpipers. Amer. Zool. 14: 185- 204. $AFRIEL, U. N. 1981. Social hierarchy among sib- lings in broods of the Oystercatcher Haematopus ostralegus. Behav. Ecol. Sociobiol. 9: 59-63. SOIKKELI, M. 1967. Breeding cycle and population dynamics in the Dunlin (Calidris alpina). Ann. Zool. Fennica 4: 158-198. SOK^L, R. R., & F. J. ROHLF. 1981. Biometry, second ed. San Francisco, Freeman and Co. V/,IS/,NEN, R. A. 1969. Evolution of the Ringed Plover (Charadrius hiaticula L.) during the last hundred years in Europe. A new computer method based on egg dimension. Ann. Acad. Sci. Fennica A. IV 149: 1-90. ß , O. HILDN, M. SOIKKELI, & S. VUOLANTO. 1972. Egg dimension variation in five wader species: the role of heredity. Ornis Fennica 40: 25-44. WARHAM, J. 1974. The Fiordland Crested Penguin Eudyptes pachyrhynchus. Ibis 116: 1-23.