Aspartic acid

ABSTRACT.-In a captive-feeding study using Herring Gulls (Larus argentatus), plasma amino-acid concentrations increased in response to an increase in dietary protein. Plasma amino-acid concentrations were also measured in wild Herring Gulls captured during incubation at eight Laurentian Great Lakes colonies. Those concentrations were used as an indicator of protein availability at those locations. Significant differences in amino acid concentrations were observed among colonies. Lower amino acid levels, particularly of the essential amino acids, were measured in gulls nesting on Lake Superior, whereas values in gulls captured on Lake Ontario and Lake Erie were greater. Those geographic differences in protein availability likely reflected spatial differences in availability of high quality prey (e.g. fish). Geographic differences in prey availability probably affected diet composition. Comparison of amino-acid levels in wild birds to reference values obtained through the captive feeding study indicated that gulls nesting on Lake Superior may have been protein limited. Colony-wide estimates of adult female body condition, intraclutch variation in egg size, and productivity were correlated with an index of plasma amino-acid concentrations. Received 30 June 2000, accepted 10 October 2001.

RESUMEN.-En un estudio en cautiverio de Larus argentatus, las concentraciones plasmaticas de aminoacidos incrementaron en respuesta a un incrementaro de proteinas en la dieta. Las concentraciones plasmaticas de aminoacidos tambien fueron medidas en individuos silvestres de L. argentatus capturados durante periodos de incubacion en ocho colonias de los Grandes Lagos. Estas concentraciones fueron usadas como indicador de la disponibilidad de proteinas en estas localidades. Se observaron diferencias significativas en las concentraciones de aminoacidos entre colonias. Los niveles de aminoacidos, particularmente de aminoacidos escenciales, fueron menores en las gaviotas nidificando en el Lago Superior y mayores en las gaviotas capturadas en el Lago Ontario y en el Lago Erie. Estas diferencias geograficas en la disponibilidad de proteinas probablemente reflejaron diferencias espaciales en la disponibilidad de presas de alta calidad (e.g. peces). Las diferencias geograficas en la disponibilidad de presas probablemente afectaron la composition de la dieta. Comparaciones de los niveles de aminoacidos de aves silvestres con valores de referencia obtenidos durante el estudio de alimentaci6n en cautiverio indicaron que las gaviotas que nidifican en el Lago Superior pueden haber estado limitadas por la disponibilidad de proteinas. Estimaciones para colonias enteras de la condition corporal de las hembras adultas, de la variaci6n en el tamano de los huevos de una misma nidada y de la productividad estuvieron correlacionadas con un indice que mide la concentration plasmatica de aminoacidos.

THE HEALTH AND reproductive success of birds can be affected by many factors (Morrison 1986). In the Laurentian Great Lakes, emphasis has been placed on the utility of fish-eating birds as indicators of effects of anthropogenic contaminants, particularly research involving Herring Gulls (Larus argentatus). During the 1970s, reproductive success of Herring Gulls and other species of fish-eating birds in the Great Lakes was poor because of exposure to environmental contaminants such as DDE, polychlorinated dibenzo-p-dioxins, polychlorinated biphenyls, and hexachlorobenzene (Gilbertson 1974, Gilbertson et al. 1991, Hebert et al. 1999a). By the late 1970s, as contaminant levels declined, reproductive success improved at most Herring Gull colonies (Allan 1991). However, in the late 1970s and early 1980s reproductive success of Herring Gulls nesting on some colonies on Lake Superior was found to be low (Allan 1991).

In general, organochlorine levels in Herring Gull eggs from Lake Superior are lower than in eggs from the other lakes (Bishop et al. 1992, Pettit et al. 1994). Therefore, it appeared that contaminants were not responsible for the observed reproductive problems. Initial indications were that Herring Gulls exhibiting poor reproductive success in parts of Lake Superior and Lake Huron may have been nutrient stressed (McNicol et al. 1985, Ewins et al. 1992, Weseloh et al. 1994).

Birds require a number of nutrients for efficient metabolism, production, thermoregulation and movement (Murphy 1996). One of the most important requirements is for adequate dietary protein enabling synthesis of proteins for maintaining body stores, for molt, and for egg production. Protein synthesis requires a source of dietary protein that can supply adequate levels of the 10 essential amino acids that cannot be synthesized by the bird and nitrogen for formation of the nonessential amino acids (Murphy 1996).

Herring Gulls are opportunistic foragers whose diet reflects their preference for certain foods and its relative availability (Pierotti 1982). Great Lakes Herring Gulls have been categorized as opportunistic piscivores (Fox et al. 1990, Belant et al. 1993, Hebert et al. 1999b). Fish may be preferred because of their high caloric and nutritional value; however, other foods may be consumed when fish are scarce. Herring Gulls nesting on Lake Superior have a lower proportion of fish in their diets than birds nesting on the lower Great Lakes (Fox et al. 1990, Hebert et al. 1999b). Lake Superior birds primarily consume anthropogenic waste (Fox et al. 1990, Hebert et al. 1999b). Inclusion of garbage in the Herring Gull diet has been found to improve breeding success in some areas (Kadlec and Drury 1968, Hunt 1972, Pons 1992, Sibly and McCleery 1983). However, evidence suggests that garbage may be inferior to more "natural" foods from a nutritional perspective and its consumption may lead to reduced reproductive success (Pierotti and Annett 1987). Assessing the value of a food as a source of protein requires information on the quantity of protein in a food and quality of that protein. Protein quality can be elucidated by examining levels of essential amino acids in a food compared with a birds' requirements during processes such as molt or egg production (Murphy and King 1992, Murphy 1994). The least available, or limiting, essential amino acid may affect the capacity of a bird to synthesize body and egg proteins (Murphy 1994).

The role of dietary differences in regulating reproductive success of Herring Gull populations within the Great Lakes has not been addressed. Past studies documenting food stress in Lake Superior Herring Gulls did not specifically examine protein availability nor did they identify endpoints that could be used to characterize effects of lowered protein availability on Herring Gull physiology and reproduction. This study examines effect of diet on plasma amino acid levels and summarizes differences in amino acid concentrations in Herring Gulls from Canadian waters of three of the Great Lakes during the 1995 and 1996 breeding seasons. Intercolony differences are examined and compared to reference values obtained through controlled feeding studies. Differences in plasma protein levels are examined in light of dietary differences among colonies. Relationships between protein availability and physiological and reproductive endpoints are discussed.

METHODS

Assessing protein availability.-Data were collected from adult Herring Gulls nesting on eight Great Lakes colonies during 1995 and 1996. Three colonies were located on Lakes Superior (Silver Islet-SIL, Marathon-MAR, Caribou Island-CAR) and Erie (Middle Sister Island-MSI, Mohawk Island-MHK, Port Colborne-PTC) and two on Lake Ontario (Scotch Bonnet Island-SBI, False Duck Island-- FDK) (Fig. 1). The Port Colborne colony was only sampled in 1995. Adults were trapped during mid to late incubation using walk-in nest traps (Mills and Ryder 1979). Time of sampling was recorded. A mean time of 7 h was spent sampling on each colony in each of the two years.

Approximately 8 mL of blood were drawn from the brachial vein of both male and female birds using a 22 gauge needle with a 10 mL heparinized vacutainer. Whole-blood samples were centrifuged in the field for 5 min at 3,200 rpm. Red blood cells and plasma were separated using serum separators and plasma was pipetted into 1.5 mL NUNC* vials and immediately frozen in liquid nitrogen.

In addition to those wild birds, samples were obtained from prefledged Herring Gull chicks kept in captivity. Twelve chicks, ~21 days of age, were captured on Scotch Bonnet Island in 1997 and transported to an outdoor holding facility at Bowmanville, Ontario. Mass of those chicks upon capture (mean +/-1 SD = 887 +/- 83 g) did not differ statistically from mass of wild adults sampled in either year (1995, 933 +/- 71 g; 1996, 957 +/- 102 g; ANOVA, P > 0.05). Approximately 18 h after their capture, a 5 mL blood sample was collected from each bird. Birds were not fed during that 18 h period. Therefore, their plasma amino acid concentrations were expected to represent a condition of short-term food deprivation. Because we do not know when chicks had last been fed prior to capture, the estimate of 18 h of food deprivation may be somewhat underestimated. After the initial blood sample was taken, each bird was fed an ad libitum protein diet. Additional 5 mL blood samples were collected from each bird after 4, 9, and 14 days. Plasma amino acid concentrations in birds from those two groups; protein-deprived birds, and birds fed protein ad libitum were used as reference values to which the wild bird amino acid levels could be compared.

Amino acid analysis.-Twenty amino acids were measured. Those included the essential amino acid: arginine (Arg), isoleucine (Ile), leucine (Leu), lysine (Lys), methionine (Met), phenylalanine (Phe), threonine (Thr), tryptophan (Trp), tyrosine (Tyr), and valine (Val). Nonessential amino acid quantified included: alanine (Ala), asparagine (Asn), aspartic acid (Asp), glutamine (Gln), glutamate (Glu), glycine (Gly), hydroxyproline (OHPro), proline (Pro), serine (Ser), and taurine (Tau). Plasma amino acid concentrations were measured using reverse-phase high performance liquid chromatography (HPLC; Waters, Millipore Ltd., Milford, Massachusetts). Briefly, 50 (mu)L of plasma were combined with 40 (mu)L of 2.5 (mu)mol/mL internal standard norleucine (Sigma Chemical Co., St. Louis, Missouri), and 1 mL of a precipitation solution (0.5% trifluroacetic acid in methanol). After vortexing, the protein was removed by centrifugation at 5,000 rpm for 5 min (Biofuge 13(TM), Heraeus Instruments, West Germany) and samples were freeze-dried overnight (Lyph@Lock 4.5; Labconco, Fisher Scientific, Nepean, Ontario). Fifty microliters of a solution of triethylamine (Aldrich Chemical Co. Inc., Milwaukee, Wisconsin), methanol and water at a concentration of 1:13 was added and the sample freeze-dried to dryness. The amino acids were derivatized for 35 min, with 20 (mu)L of a 1:1:7:1 solution of water, triethylamine, methanol, and phenylisothiocyanate (Pierce, Rockford, Illinois), (Bidlingmeyer et al. 1984), and then freeze-dried overnight. The amino acid derivatives were resuspended in 200 (mu)L of a solution containing 95% phosphate buffer (5 mM Na^sub 2^HPO^sub 4^adjusted to pH 7.4 with 10% phosphoric acid) and 5% acetonitrile (HPLC grade; Fisher Scientific, Nepean, Ontario). Samples were then centrifuged at 5,000 rpm for 5 min.

Ten microliter aliquots of the derivatized samples were injected onto a C^sup 18^ reversed-phase column, maintained at 46 deg C and eluted using a convex binary gradient. Buffer A contained 70 mM sodium acetate (BDH Inc., Toronto, Ontario), and 2.5% acetonitrile, adjusted to pH 6.55 with concentrated acetic acid. Buffer B consisted of a 45:40:15 ratio of acetonitrile, water, and methanol. Amino acid derivatives were detected at a wavelength of 254 nm and processed through a datastation (model 840; Waters, Millipore Ltd., Milford, Massachusetts).

Amino acid concentrations were calculated as follows: amino acid ((mu)mol/L) = (amino acid peak area/norleucine peak area) x CF x RF, where CF (concentration factor) = 250 (mu)mol/L, RF (response factor) = norleucine peak area/amino acid peak area.

Assessing food quality.-To assess the relative quality of different foods to support egg formation, we examined amino acid concentrations and patterns in a wide variety of foods and compared them with those measured in avian eggs. Those foods that had amino acid patterns similar to egg patterns were expected to be more suitable in supporting amino acid requirements of birds during egg production (Murphy 1994). Data concerning amino acid levels in all foods were obtained from Food and Agriculture Organization (1970) and those data represented mean values for each food calculated using many replicate samples. For the 19 food types included in our analysis, a total of 306 individual samples were analyzed to calculate average amino acid composition of each food. Amino acid patterns in foods originating from anthropogenic sources (i.e. garbage) were estimated from patterns observed in various starches (potato, rice, corn, wheat) and meats (beef, chicken, lamb, pork). Natural foods for which amino acid patterns were also examined included fish (i.e. order Clupeiformes, suborder Clupeoidei [e.g. herring], Clupeiformes Salmonoidei [e.g. smelt], Cypriniformes [e.g. carp], Perciformes [e.g. perch]), crustaceans, small mammals (rats), and terrestrial invertebrates (caterpillars) (Food and Agriculture Organization 1970). Amino acid levels and patterns in those foods were compared with those found in chicken and duck eggs (Food and Agriculture Organization 1970). Although no Herring Gull eggs were analyzed, Murphy (1994) found that amino acid composition of eggs from five avian orders (Anseriformes, Galliformes, Columbiformes, Psittaciformes, Passeriformes) were very similar. Therefore, we expected that foods with amino acid patterns most closely resembling those in eggs of the bird species examined here would also reflect suitability of those foods for production of Herring Gull eggs.

Assessing effects of dietary protein on physiology and reproduction.-Variation in body size must be considered when comparing indices of body condition in birds (see Blem 1990). The first principal component (PCI) of a principal components analysis (PCA) has been found to be a good predictor of overall body size in birds (Rising and Somers 1989, Freedman and Jackson 1990). In this study, PC1 scores from a PCA using the correlation matrix of morphological variables were used as a measure of structural size. Body mass and six morphometrics were taken from all birds blood sampled. Measures of body size included (1) head and bill length, maximum distance from bill tip to posterior extremity of the occipital process; (2) bill depth, with mandibles closed, minimum depth posterior to the gonyx; (3) wing chord, with the wing flattened and flexed at the wrist, from wrist to tip of wing; (4) tarsus length, from pit at the junction of the tibiotarsus and tarsometatarsus to the distal end of tibiotarsus; (5) keel length, distance between the notch at the furculum and posterior-most point along the median of the sternum; and (6) sternum diagonal, length from the midpoint of the posterior end of the sternum to distal end of either clavicle. All measurements were taken to the nearest 0.1 mm with calipers except wing length, which was measured to the nearest millimeter with a flat ruler.

An ANCOVA was used to examine whether differences between sexes existed with respect to the relationship between body mass and PCd. If significant differences were observed (1995 data), separate sexspecific regression relationships were used to calculate residuals. If differences were not detected, then a common regression was used (1996 data). Residuals were the difference between measured body mass and that predicted from the mass to body size regression. Those residuals were used as an indicator of body condition. Body mass corrected for structural size has been recognized as a condition index because changes in body weight are largely due to changes in total body lipid (Chappel and Titman 1983, Whyte and Bolen 1984). Only the results for females are reported here. The sex of each bird was determined using morphological characteristics (Fox et al. 1981).

Egg measurements from the sampled adult's nest were collected allowing egg volumes to be calculated using the formula 0.000476 x length x breath^sup 2^ (Harris 1963). Only birds with three-egg clutches were included in those measurements. Clutch volume was total volume of the three eggs in the clutch. Intraclutch variation in egg size represents difference in volume (cubic centimeters) between largest and smallest eggs in a clutch. Overall estimates of colony productivity, measured as number of 21 day-old chicks per apparently occupied nest (AON; Mineau and Weseloh 1981) were obtained during a later visit to each colony. At that time, colonies were thoroughly surveyed and number of chicks recorded. Twentyone days is an arbitrary but commonly used age after which posthatch mortality is minimal (see Mineau and Weseloh 1981).

Statistical analysis.-Plasma amino acid concentrations in Herring Gulls were not normally distributed. For that reason, randomization methods (permutation tests) were used to test for significant differences in amino acid concentrations between sexes and among groups. Between-sex comparisons were performed for each colony and year separately. Among-- group comparisons were completed using data from the wild birds along with results from the two groups in the captive feeding study. For the captive birds fed protein ad libitum, mean amino acid concentrations were calculated for each bird from blood samples collected after 4, 9, and 14 days. Those mean values were used in the intergroup comparisons. Among-group comparisons were performed separately for the 1995 and 1996 data.

Permutation tests (see Manly 1997) involved combining amino acid results from all groups for individual years and then randomly permuting those observations into subgroups of the original sample size. Test statistics resulting from 5,000 randomizations were compiled to form a frequency distribution. The fraction of the test statistics that equaled or exceeded the observed difference between means was determined. That fraction was the probability of obtaining the observed result.

Principal components analysis (StatSoft 1995), using correlation matrices, was used to reduce dimensionality of amino acid data and facilitate their interpretation. A PCA was conducted on all of the amino acid data (20 amino acids) from the wild Herring Gulls sampled in 1995 and 1996 in addition to the Herring Gull chicks sampled in captivity. Principal components analysis was also used to examine differences in amino acid levels and patterns in foods that Herring Gulls consume. That analysis included 15 amino acids. Data for Trp, Asn, Gln, OHPro, and Tau were not available (see Food and Agriculture Organization 1970).

Spearman correlations were used to examine the relationship between amino acid concentrations and time of sampling at each colony for each year. Spearman correlations were also used to examine the relationship between plasma amino acid concentrations, as summarized using PCA, and female body condition, clutch volume, intraclutch variation in egg size, and productivity.

RESULTS

Assessing protein availability.-Plasma amino acid concentrations in the captive Herring Gull chicks increased following administration of an ad libitum protein diet. However, for individual amino acids, differences were rarely statistically significant (Tables 1 and 2).

Results from permutation tests indicated that there were no significant differences in plasma amino acid concentrations between wild male and female Herring Gulls at any colony (P > 0.05). Therefore, results from males and females were pooled to examine intergroup differences. There was no relationship between time of sampling and amino acid concentrations in birds sampled at any colony in either year (P > 0.05), suggesting that investigator disturbance was not an important factor affecting plasma amino acid concentrations.

Differences in plasma amino acid concentrations were observed among Herring Gulls from different colonies and from the two reference groups (Tables 1 and 2). In general, amino acid concentrations were lower in gulls from colonies on Lake Superior than from colonies on lakes Erie or Ontario. That pattern was more apparent for essential amino acids (Table 1) than nonessential amino acids (Table 2).

Because of the number of amino acids analyzed and the complexity of the intergroup comparisons, PCA was used to identify general patterns in the results. The PCA of all samples indicated that PC1 provided an overall indication of plasma amino acid concentration. The first principal component explained 28% of the variation in the data and all of the amino acids loaded positively on this axis (Table 3). Although all amino acids loaded positively on PC1, it was evident that the most important amino acids contributing to the differentiation of groups were the essential amino acid: Leu, Thr, Val, Met, and Phe. Of the nonessential amino acid, only Pro and OHPro exhibited high loadings. OHPro is synthesized from Pro (Stevens 1996). Levels of these amino acids are shown in more detail along with an indication of intergroup differences (Fig. 2).

Scores from PC1 provided a summary of overall differences in plasma amino acid concentrations with particular emphasis on essential amino acids noted above. Using randomization procedures, intergroup differences in this index of plasma amino acid concentrations were identified (hereafter referred to as "plasma amino acid index"; Fig. 3). That analysis was performed using data from both 1995 and 1996 in addition to the reference groups from the captive feeding study. It is evident that in the captive feeding study the protein deprived birds had significantly lower plasma amino acid concentrations than birds fed the ad libitum protein diet. It is also evident that, in general, birds from Lake Superior exhibited lower plasma amino acid concentrations than birds from lakes Ontario and Erie. The birds from the two Lake Superior colonies at Marathon and Caribou Island were the only birds to show consistently lower plasma amino acid concentrations than the captive birds fed protein ad libitum.

Assessing food quality.-Principal components analysis revealed differences in amino acid levels and patterns in foods that Herring Gulls consume (Fig. 4). The first principal component was an indicator of differences in overall amino acid concentrations in the various food types and in avian eggs. The first principal component explained 70% of the variance in the data. Although all of the amino acids loaded positively on PC1, the essential amino acids were of greatest importance (Table 4). The first principal component primarily served to separate animal from plant foods. Animal foods had higher amino acid concentrations. Within animal foods, fish had consistently high amino acid levels. The second principal component summarized 10% of variance in the data and emphasized differences in amino acid patterns among food types and avian eggs. The amino acid that loaded most positively on PC2 was Met (Table 4). It is evident that the samples that had the highest relative Met concentrations were the avian eggs (Fig. 4). Of the foods with high overall amino acid levels (i.e. animal foods), the fish samples showed the greatest Met levels and were most similar to the egg samples.

Assessing effects of dietary protein on physiology and reproduction.-No correlation was found between the plasma amino acid index and colony-- specific estimates of clutch volume (r^sub s^ = 0.09, P = 0.76). After adjusting clutch volume for female size, there was still no significant correlation with the plasma amino acid index. Significant correlations were observed between the plasma amino acid index and adult female body condition (Fig. 5a), intraclutch variation in egg size (Fig. 5b), and productivity (Fig. 5c).

DISCUSSION

Results from the captive feeding study indicated that overall plasma amino acid concentrations increased significantly with increased dietary protein intake. Similarly, Jeffery et al. (1985) measured plasma levels of 26 amino acids in Herring Gull chicks kept in captivity. After four days of starvation, levels of 23 of the amino acids had declined, 13 significantly. The greater number of statistically significant declines observed by Jeffery et al. (1985) compared to the results from the captive feeding study described here probably reflected the longer period of food deprivation in the earlier study. Regardless, results from both studies suggested that we could interpret higher plasma amino acid concentrations in wild birds as being indicative of enhanced dietary protein availability. Those studies also indicated that plasma amino acid levels reflected recent dietary protein availability with changes in plasma levels coinciding with dietary changes on the order of days.

Intercolony differences in plasma amino acid concentrations indicated that there may have been differences among colonies in terms of protein availability. Compared to plasma amino acid concentrations in captive birds fed an ad libitum protein diet, birds from two of the three Lake Superior colonies had significantly lower amino acid levels similar to those in captive birds that had experienced short-term protein deprivation. Although Herring Gull chicks were used in the captive feeding study, the mass of the chicks was not significantly different from wild adults. Amino acids may be used differently in chicks than in adults because of formation of tissues during chick growth. For example, the demand for proline is high during early chick growth (Baker 1991) and feather development compared to other life stages (summarized in Murphy 1996). If rapid growth had been occurring in the captive chicks, we would have expected that plasma proline levels in captive birds to have been lower than in wild adults. In 1996, that was not the case and in 1995 only birds from False Duck Island on Lake Ontario had significantly greater total proline concentrations than captive chicks on the ad libitum protein diet (Fig. 2). Because most chick growth had occurred prior to their being brought into captivity, potential differences in amino acid use patterns between captive chicks and wild adults were probably minimized.

Although not always statistically significant, overall plasma amino acid concentrations were lower in birds from the two colonies in central and eastern Lake Superior than in birds from the lower Great Lakes, particularly Lake Ontario. This indicates that the diet of Herring Gulls nesting on Lake Superior may have been deficient in amino acids compared to diets of Herring Gulls nesting on colonies on lakes Ontario and Erie. The intermediate levels of amino acids generally observed in birds nesting on Lake Erie were somewhat surprising. Historically, Lake Erie has been the most productive Great Lake (Neilson et al. 1995) and we would have expected that protein availability in the form of fish would have been high. However, large changes in the Lake Erie ecosystem have occurred in recent years and may, in part, reflect the influence of introduced dreissenid mussels on that ecosystem (Leach 1993, Nicholls and Hopkins 1993, Howell et al. 1996). There are indications that dietary habits of Herring Gulls on Lake Erie have changed through time and those changes may have reflected a decline in fish availability (Hebert et al. 1999b, 2000). Future research should continue to monitor diet composition in Lake Erie fish-eating waterbirds as an indication of further change in fish or protein availability.

In most of the Great Lakes, fish are the primary protein source for Herring Gulls (Fox et al. 1990, Ewins et al. 1994, Hebert et al. 1999b). Fish availability to Herring Gulls in different regions of the Great Lakes varies greatly and may depend on aquatic production, human fishing activities (Hebert et al. 1999b), or both. For example, in Lake Superior, one of the primary prey fish species is the rainbow smelt (Osmerus mordax). From 1989 to 1996, annual estimates of their abundance were 2-9 x greater in the western part of the lake versus the eastern portion (Great Lakes Fishery Commission 1997). That, in part, might explain the relatively higher plasma amino acid concentrations in birds from the Silver Islet colony relative to values observed in birds from the other two Lake Superior colonies (see Fig. 3). In general, however, diets of Lake Superior gulls contained a much greater proportion of garbage than did diets of lower Great Lakes gulls (Hebert et al. 1999b).

From analysis of potential Herring Gull food items, it was evident that anthropogenic sources of foods, particularly plant-based foods, contained lower amino acid concentrations than fish. In addition, amino acid patterns observed in fish, particularly with respect to their relatively high Met levels, most closely resembled those in avian eggs. The quality of a food as a protein source reflects both quantity of amino acids in the food and degree to which that food meets the relative amino acid requirements for processes such as egg formation. For egg production, foods with high overall amino acid concentrations and similar amino acid patterns to eggs would be of the best quality. Therefore, foods situated in the northeast quadrant of Figure 4 would represent the highest quality foods for egg production. Those foods are predominantly fish containing high overall amino acid concentrations and their amino acid patterns, particularly with respect to a high relative abundance of Met, resemble those in eggs. Murphy (1994) predicted that amino acids that may be most limiting during egg production are the sulphur-containing amino acids, that is, Met and cysteine (not measured in this study). Therefore, it is reasonable to conclude that the diet of gulls consuming garbage will be inferior to that of those eating fish. These results suggest that garbage is of poor nutritional value and support the conclusions of Pierotti and Annett (1987).

Lower protein availability for gulls nesting on Lake Superior could be partly responsible for the lower reproductive success observed on that lake since the late 1970s and early 1980s. To examine that issue and the effect of diet quality on adult physiological condition, annual colony-wide estimates of female body condition, egg volume, and fledging success were correlated with plasma amino acid index. Because different physiological and reproductive processes require specific amino acids in different proportions, for example egg production requires high Met, chick growth requires high Pro, no one amino acid would be expected to correlate with all physiological and reproductive endpoints. However, plasma amino acid index provides an integrative measure of amino acid availability, with particular emphasis on the essential amino acids. Use of this index may provide a more universal indicator of effects of protein availability on body condition and reproduction.

Following ingestion, dietary proteins are hydrolyzed by proteolytic enzymes in the gastrointestinal tract. That results in the release of free amino acids that are then absorbed into the blood from the small intestine and transported to the liver (Stevens 1996). Those free amino acids, in conjunction with amino acids released via turnover of endogenous protein stores, are used as building blocks for synthesis of new protein (Murphy 1996). The importance of pectoral muscle protein as a source of endogenous amino acids for protein destined for egg formation has been documented in the Lesser Black-backed Gull (Larus fuscus; Houston et al. 1983). Adults with lower proportions of protein in their diet may be hampered in their ability to synthesize body protein. That could lead to a reduction in adult condition if body reserves were depleted during egg formation. The statistically significant relationship between plasma amino acid index and female body condition indicated that differences in protein availability may have affected the ability of female Herring Gulls to synthesize body reserves.

Costs associated with egg formation in gulls are thought to be significant (Monaghan and Nager 1997). The importance of an adequate supply of dietary protein during egg formation has also been documented in larids (Hiom et al. 1991, Bolton et al. 1992). Feeding studies have shown that increasing the amount of protein in the diet of laying chickens will result in increases in egg size (Leeson and Summers 1997). Similar increases in egg size were also observed when the proportion of Met in the laying female's diet was increased (Calderon and Jensen 1990, Waldroup and Hellwig 1995). The significant negative correlation between the plasma amino acid index and amount of intraclutch variation in egg size found in this study indicated that, at colonies where protein may have been limited, resources for egg formation may have been reduced. Kilpi et al. (1996) concluded that in three-egg Herring Gull clutches, differences in egg size between the first (the aegg) and last laid egg (the c-egg) reflected feeding conditions during egg laying. In larids, the size of the a-egg is generally greater than the c-egg but Kilpi et al. (1996) pointed out that the degree of difference in intraclutch egg size varied among colonies. They hypothesized that the increased difference in size between the aand c-eggs may have reflected poorer feeding conditions. Herring Gulls likely rely on both endogenous and exogenous lipid and protein stores for egg formation (Drent and Daan 1980, Hario et al. 1991). The time-frame associated with clutch initiation and completion is relatively fixed, with eggs being laid at approximately two day intervals (Drent 1970). During that period, protein stores for formation of the last laid egg may become critical, particularly under poor feeding conditions (Kilpi et al. 1996).

In this study, we used a similar approach in our calculation of the difference between the largest and the smallest egg within three-egg Herring Gull clutches. Because we did not know the sequence of laying, we could not designate egg position; however, in most studies the c-egg is smallest (Parsons 1970, 1975; Davis 1975; Spaans et al. 1987; Kilpi et al. 1996; Nager et al. 2000). With decreasing protein (i.e. fish) availability, resources (including Met) available for formation of the smallest egg (the c-egg) may have been more constrained than for the larger a- and b-eggs. That may have resulted in the greater relative difference in size between largest and smallest eggs that we observed on Lake Superior colonies. Kilpi et al. (1996) observed that reproductive success was lower when the degree of difference in egg size between a- and c-eggs was greater. When that difference was greater than ~9%, productivity was reduced. Our estimates of intraclutch egg size variation are not directly comparable with values summarized in Kilpi et al. (1996) because they are based upon differences between largest and smallest eggs within a clutch. That results in our estimates of within clutch egg size variation being greater than those summarized by Kilpi et al. (1996). However, it is interesting to note that at colonies where intraclutch egg size variation exceeded 12%, productivity was low and similar to or below the level of productivity that is thought to be required to maintain a stable population (0.8-- 1.4 chicks per nest; Kadlec and Drury 1968, Kadlec 1976). For routine environmental monitoring, the degree of difference between size of the largest and smallest eggs can be measured much more easily than difference in size between the a- and c-eggs. That is because for the latter measurement, detailed observations on laying chronology are required, necessitating repeated visits to the same colony over a period of days. Our endpoint has the potential to be readily used in field surveys to provide an initial indication of dietary stress in colonial waterbirds that lay multiple-egg clutches and may be calculated using data obtained during one visit.

Reduced egg size is thought to negatively affect survival of Herring Gull chicks (Parsons 1970) but that conclusion has been challenged (Davis 1975). In gulls and terns, differences in chick survival may be more affected by compositional changes in the amount of lipid or protein in the eggs than by overall size of the egg (Nisbet 1978, Nager et al. 2000). In this study, the significant relationship between the plasma amino acid index and colony productivity indicated that dietary stress may have played some role in the reduced reproductive success of Herring Gull colonies on Lake Superior. One mechanism for that to be manifested is through the potential effect of small egg size and low egg protein-lipid content on chick survival. However, as emphasized in Kilpi et al. (1996) we do not suggest that there is a direct cause and effect relationship between relative egg size and productivity. Instead, a number of other factors related to poor feeding conditions, such as low rates of chick provisioning and reduced chick guarding by parents, may have been more important mechanisms linking dietary stress to low productivity (see Heaney and Monaghan 1995, Monaghan et al. 1998).

This study has developed sampling and statistical techniques that can be used to evaluate protein availability in the Herring Gull diet. Results from this work will provide useful comparative data for other studies attempting to evaluate nutritional stress in larids. As we try to understand the relative importance of multiple stressors on wildlife, we must develop the means to assess their potential effects. Inclusion of a broader number of endpoints, indicative of multiple stressors and potential ecosystem change, will enhance the utility of ongoing biomonitoring programs such as the Herring Gull Monitoring Program.

ACKNOWLEDGMENTS

The authors thank K. Williams (CWS, NWRC) for his expert assistance in the field. K. Brown collected the blood samples from Port Colborne in 1995. Staff at the AviCare Wild Bird Care Centre in Bowmanville, Ontario cared for the captive gulls, and H. Pittel and M. Metcalf took blood samples from those birds. G. Law and K. Shoveller (University of Guelph) performed the amino acid analyses. B. Collins (CWS, NWRC) supplied the program to conduct the randomization tests. G. Fox, K. Keenleyside, and G. Sprules provided suggestions to improve the manuscript. Environment Canada's Great Lakes Action Plan provided funds for this study.

LITERATURE CITED

ALLAN, R.J. 1991. Toxic chemicals in the Great Lakes and associated effects. Environment Canada, Toronto. Ontario.

BAKER, D. H. 1991. Partitioning of nutrients for growth and other metabolic functions: Efficiency and priority considerations. Poultry Science 70:1797-1805.

BELANT, J. L., T. W. SEAMANS, S. W. GABREY, AND S. K. ICKES. 1993. Importance of landfills to nesting Herring Culls. Condor 95:817-830.

BISHOP, C. A., D. V. WESELOH, N. B. BURGESS, R. J. NORSTROM, R. TURLE, AND K. A. LOGAN. 1992. An atlas of contaminants in eggs of colonial fisheating birds of the Great Lakes (1970-1988), vols. I and II. Technical Report Series, no. 153. Canadian Wildlife Service, Ottawa, Ontario.

BLEM, C. R. 1990. Avian energy storage. Current Ornithology 7:59-113.

BOLTON, M., D. HOUSTON, AND P. MONAGHAN. 1992. Nutritional constraints on egg formation in the Lesser Black-backed Gull: An experimental study. Journal of Animal Ecology 61:521-532.

CALDERON, V. M., AND L. S. JENSEN. 1990. The requirement for sulfur amino acid by laying hens as influenced by protein concentration. Poultry Science 69:34-944.

CHAPPEL, W. A., AND R. D. TITMAN. 1983. Estimating reserve lipids in Greater Scaup (Aythya marila) and Lesser Scaup (A. affinis). Canadian Journal of Zoology 61:35-38.

DAVIS, J. W F. 1975. Age, egg size and breeding success of the Herring Gull Larus argentatus. Ibis 117:460-472.

DRENT, R. 1970. Functional aspects of incubation in the Herring Gull. Pages 1-132 in The Herring Gull and its Egg (G. P. Baerends and R. H. Drent, Eds.). Behaviour Supplement, no. 17.

DRENT, R. H., AND S. DAAN. 1980. The prudent parent: Energetic adjustments in avian breeding. Ardea 68:225-252.

EWINS, P. J., D. V. WESELOH, AND P. MIN EAU. 1992. Geographical distribution of contaminants and productivity measures of Herring Gulls in the Great Lakes: Lake Huron 1980. Journal of Great Lakes Research 18:316-330.

EWINS, P. J., D. V. WESELOH, J. H. GROOM, R. Z. DoBOS, AND P. MINEAU. 1994. The diet of Herring Gulls (Larus argentatus) during winter and early spring on the lower Great Lakes. Hydrobiologia 279/280:39-55.

FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS. 1970. Amino-acid content of foods and biological data on proteins. Food and Agriculture Organization, Rome, Italy.

Fox, G. A., L. J. ALLAN, D. V. WESELOH, AND P. MINEAU. 1990. The diet of Herring Gulls during the nesting period in Canadian waters of the Great Lakes. Canadian Journal of Zoology 68:10751085.

Fox, G. A., C. R. COOPER, AND J. P. RYDER. 1981. Predicting the sex of Herring Gulls by using exter

nal measurements. Journal of Field Ornithology 52:1-9.

FREEDMAN, S., AND W. M. JACKSON. 1990. Univariate metrics are not adequate to measure avian body size. Auk 107:69-74.

GILBERTSON, M. 1974. Pollutants in breeding Herring Gulls in the lower Great Lakes. Canadian Field-- Naturalist 8:273-280.

GILBERTSON, M., T. KUBIAK, J. LUDWIG, AND G. Fox. 1991. Great Lakes embryo mortality, edema, and deformities syndrome (GLEMEDS) in colonial fish-eating birds: Similarity to chick-edema disease. Journal of Toxicology and Environmental Health 33:455-520.

GREAT LAKES FISHERY COMMISSION. 1997. Population status and trends for Lake Superior forage fishes, 1978-1996. Lake Superior Committee Meeting Report. Ann Arbor, Michigan.

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. 1963. Aspects of the breeding biology of the gulls Larus argentatus, L. focus and L. marinus. Ibis 106:432-456.

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, Series B 261:361-365.

HEBERT, C. E., K. A. HOBSON, AND J. L. SHUTT. 2000. Changes in food web structure affect rate of PCB decline in Herring Gull (Larus argentatus) eggs. Environmental Science and Technology 34:16091614.

HEBERT, C. E., R. J. NORSTROM, AND D. V. WESELOH. 1999a. A quarter century of environmental surveillance: The Canadian Wildlife Service's Great Lakes Herring Gull Monitoring Program. Environmental Reviews 7:147-166.

HEBERT, C. E., J. L. SHUTT, K. A. HOBSON, AND D. V. WESELOH. 1999b. Spatial and temporal differences in the diet of Great Lakes Herring Gulls (Larus argentatus): Evidence from stable isotope analysis. Canadian Journal of Fisheries and Aquatic Sciences 56:323-338.

HIOM, L., M. BOLTON, P. MONAGHAN, AND D. WoRRALL. 1991. Experimental evidence for food limitation of egg production in gulls. Ornis Scandinavica 22:94-97.

HOUSTON, D. C., P. J. JONES, AND R. M. SIBLEY. 1983. The effect of female body condition on egg laying in Lesser Black-backed Gulls Larus fuscus. Journal of Zoology (London) 200:509-520.

HOWELL, E. T., C. H. MARVIN, R. W. BILYEA, P. B. KAUSS, AND K. SOMERS. 1996. Changes in environmental conditions during Dreissena colonization of a monitoring station in eastern Lake Erie. Journal of Great Lakes Research 22:744756.

HUNT, G. L. 1972. Influence of food distribution and human disturbance on the reproductive success of Herring Gulls. Ecology 53:1051-1061.

JEFFREY, D. A., D. B. PEAKALL, D. S. MILLER, AND G. R. HERZBERG. 1985. Blood chemistry changes in food-deprived Herring Gulls. Comparative Biochemistry and Physiology A 81:911-913.

KADLEC, J. A. 1976. A re-evaluation of mortality rates in adult Herring Gulls. Bird-Banding 47:8-12. KADLEC, J. A., AND W. H. DRURY. 1968. Structure of

the New England Herring Gull population. Ecology 49:644-676.

KILPI, M., L. HILLSTROM, AND K. LINDSTROM. 1996.

Egg-size variation and reproductive success in the Herring Gull Larus argentatus: Adaptive or constrained size of the last egg? Ibis 138:212217.

LEACH, J. H. 1993. Impacts of the zebra mussel (Dreissena polymorpha) on water quality and fish spawning reefs in western Lake Erie. Pages 381397 in Zebra Mussels: Biology, Impacts, and Control (T. E Nalepa and D. W. Schloesser, Eds.). Lewis Publishers, Boca Raton, Florida.

LEESON, S., AND J. D. SUMMERS. 1997. Commercial Poultry Nutrition, 2nd ed. University Books, Guelph, Ontario.

MANLY, B. F J. 1997. Randomization, Bootstrap and Monte Carlo Methods in Biology, 2nd ed. Chapman and Hall, New York.

McNICOL, M. K., J. STRUGER, AND D. V. WESELOH. 1985. Foods and feeding rates in a small colony of cannibalistic Herring Gulls. Colonial Waterbirds Group Newsletter 9:41.

MILLS, J. A., AND J. P. RYDER. 1979. Trap for capturing shore and seabirds. Bird-Banding 50:121-123. MINEAU, P., AND D. V. C. WESELOH. 1981. Low-dis

turbance monitoring of Herring Gull reproductive success on the Great Lakes. Colonial Waterbirds 4:138-142.

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, Series B 259:189-191.

MORRISON, M. L. 1986. Bird populations as indicators of environmental change. Current Ornithology 3:429-451.

MURPHY, M. E. 1994. Amino acid compositions of avian eggs and tissues: Nutritional implications. Journal of Avian Biology 25:27-38.

MURPHY, M. E. 1996. Nutrition and metabolism. Pages 31-60 in Avian Energetics and Nutritional Ecology (C. Carey, Ed.). Chapman and Hall, New York.

MURPHY, M. E., AND J. R. KING. 1992. Energy and nutrient use during moult by White-crowned Spar

rows Zonotrichia leucophrys gambelii. Ornis Scandinavia 23:304-313.

NAGER, R. G., P. MONAGHAN, AND D. C. HOUSTON. 2000. Within-clutch trade-offs between the number and quality of eggs: Experimental manipulations in gulls. Ecology 81:1339-1350.

NEILSON, M., S. L'ITALIEN, V. GLUMAC, D. WILLIAMS, AND P. BERTRAM. 1995. Nutrients: Trends and system response. 1994 State of the Lakes Ecosystem Conference. EPA 905-R-95-015, U.S. Environmental Protection Agency, Chicago.

NICHOLLS, K. H., AND G. J. HOPKINS. 1993. Recent changes in Lake Erie (north shore) phytoplankton: Cumulative impacts of phosphorus loading reductions and the zebra mussel introduction. Journal of Great Lakes Research 19:637-647.

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. This 120:207-215.

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

PARSONS, J. 1975. Asynchronous hatching and chick mortality in the Herring Gull Larus argentatus. Ibis 117:517-520.

PETTIT, K. E., C. A. BISHop, D. V WESELOH, AND R. J. NORSTROM. 1994. An atlas of contaminants in eggs of colonial fish-eating birds of the Great Lakes (1989-1992), vols. I and II. Technical Report Series, no. 193. Canadian Wildlife Service, Ottawa, Ontario.

PIEROTTI, R. 1982. Habitat selection and its effect on reproductive output in the Herring Gull in Newfoundland. Ecology 63:854-868.

PIEROTTI, R., AND C. A. ANNETT. 1987. Reproductive consequences of dietary specialization and switching in an ecological generalist. Pages 417

442 in Foraging Behavior (A. C. Kamil, J. R. Krebs, and H. R. Pulliam, Eds.). Plenum Publishing, New York.

PONS, J. 1992. Effects of changes in the availability of human refuse on breeding parameters in a Herring Gull Larus argentatus population in Brittany, France. Ardea 80:143-150.

RISING, j. D., AND K. M. SOMERS. 1989. The measurement of overall body size in birds. Auk 106:666674.

SIBLY, R. M., AND R. H. MCCLEERY. 1983. Increase in weight of Herring Gulls while feeding. Journal of Animal Ecology 52:35-50.

SPAANS, A. L., A. N. DE WIT, AND M. A. VAN VLAARDINGEN. 1987. Effects of increased population size in Herring Gull on breeding success and other parameters. Studies in Avian Biology 10: 57-65.

STATSOFT. 1995. Statistical Procedures Manuals, vols. 1-5. StatSoft Inc., Tulsa, Oklahoma. STEVENS, L. 1996. Avian Biochemistry and Molecular

Biology. Cambridge University Press, Cambridge, United Kingdom.

WALDROUP, P. W, AND H. M. HELLWIG. 1995. Methionine and total sulfur amino acid requirements influenced by stage of production. Journal of Applied Poultry Research 3:1-6.

WESELOH, D. V C., P. J. EWINS, J. STRUGER, P. MINEAU, AND R. J. NORSTROM. 1994. Geographical distribution of organochlorine contaminants and reproductive parameters in Herring Gulls on Lake Superior in 1983. Environmental Monitoring and Assessment 29:229-251.

WHYTE, R. J., AND E. G. BOLEN. 1984. Variation in winter fat depots and condition indices of Mallards. Journal of Wildlife Management 48:13701373.

Associate Editor: C. Blem

CRAIG E. HEBERT,1,3 J. LAIRD SHUTT,1 AND RON O. BALL2

1Environment Canada, Canadian Wildlife Service, National Wildlife Research Centre, Hull, Quebec K1A 0H3, Canada; and

2Department of Agricultural, Food, and Nutritional Sciences, University of Alberta, Edmonton, Alberta T6G 2P5, Canada

3 E-mail: craig.hebert@ec.gc.ca

Copyright American Ornithologists' Union Jan 2002
Provided by ProQuest Information and Learning Company. All rights Reserved