Variability of Atlantic Coast Striped Bass Egg Characteristics

North American Journal of Fisheries Management 23:558–572, 2003 q Copyright by the American Fisheries Society 2003

Variability of Atlantic Coast Striped Bass Egg Characteristics LAUREN L. BERGEY1 Department of Biology, East Carolina University, Greenville, North Carolina 27858-4353, USA

ROGER A. RULIFSON* Department of Biology and Institute for Coastal and Marine Resources, East Carolina University, Greenville, North Carolina 27858-4353, USA

MARGIE L. GALLAGHER School of Human Environmental Sciences and Institute for Coastal and Marine Resources, East Carolina University, Greenville, North Carolina 27858-4353, USA

ANTHONY S. OVERTON Institute for Coastal and Marine Resources, East Carolina University, Greenville, North Carolina 27858-4353, USA Abstract.—Eggs of striped bass Morone saxatilis were collected from broodfish at seven hatcheries and one wild population representing nine watersheds from Georgia to Canada to determine the relationship between watershed type and egg characteristics, including density, diameter, oil globule size, surface : volume ratio, and lipid content. These populations represented an inland freshwater lake (Lake Lanier), upland-dominated (high-physical-energy) freshwater rivers (Dan and upper Roanoke [Staunton] rivers), estuarine-influenced (low-physical-energy) coastal rivers (Savannah, Pamunkey, Choptank, and Nanticoke rivers), an upland tidal-bore river (Shubenacadie River), and an upland tidal river (Miramichi River). Water quality parameters varied among hatchery locations. Water hardening of eggs occurred within 2.5 h of fertilization. Egg diameter and relative oil globule size did not differ significantly under ambient hatchery and controlled water quality conditions. However, eggs from different watersheds differed significantly in several aspects. Eggs from highphysical-energy watersheds were heavier and larger, and had smaller oil globule sizes, smaller surface : volume ratios, and larger amounts of saturated and monounsaturated fatty acids in both the neutral lipid and phospholipid fractions than eggs from slower-moving watersheds. Egg characteristics did not vary with latitude. Differences in egg characteristics likely are the result of population adaptations to the watershed, but the manner in which these adaptations were produced was not determined. Our study provides evidence that striped bass stock restoration and enhancement programs should use native broodfish for hatchery production and should return progeny to the natal watershed.

The striped bass Morone saxatilis is an anadromous species with a natural range along the eastern coast of North America from northern Florida to the St. Lawrence Estuary, Canada, and along the Gulf of Mexico from western Florida to Texas

* Corresponding author: [email protected] 1 Current address: Rutgers, the State University of New Jersey, Biology Department, 101 Warren Street, Newark, New Jersey 07102, USA. Received August 2, 2001; accepted September 4, 2002

(Setzler et al. 1980). Striped bass were successfully introduced to the San Francisco Bay region of the Pacific Coast in 1879 with 150 yearling striped bass from the Navisink River, New Jersey, and in 1881 with 300 yearling striped bass from the Shrewsbury River, New Jersey (Smith 1896). Currently, the West Coast range for the species extends from Ensenada, Mexico, to British Columbia (Forrester et al. 1972). For decades, many natural populations of the U.S. eastern and Gulf coasts were stocked with hatchery-reared fish for stock enhancement or stock restoration purposes, where-

558

559

VARIABILITY OF STRIPED BASS EGGS

as Canadian populations have not been stocked (Rulifson and Laney 1999). Until recently, hatcheries used broodfish from convenient watershed locations, so progeny from one population were often stocked in multiple (nonnative) watersheds. The possible effect of this practice on early life stage survival and reproductive success has not been addressed. A variant of this practice is the stocking of hybrid striped bass (striped bass 3 white bass M. chrysops) in watersheds containing native striped bass populations. A 9-year study by Patrick and Moser (2001) reported that hybrid striped bass stocked in the Cape Fear watershed, North Carolina, are faring better than the native population, and the authors urged termination of this practice by the state freshwater fisheries agency. Striped bass spawn from late March until late June, depending on latitude and ambient water temperatures (Setzler et al. 1980). Some populations spawn in relatively low-physical-energy (lowgradient) coastal rivers with tidal influences, whereas others spawn farther upstream beyond tidal influence under higher-physical-energy (higher-gradient) conditions characteristic of upland watersheds. One population has even adapted to spawn in a tidal-bore river typified by harsh and sudden changes in temperature, salinity, and river flow direction (Rulifson and Tull 1999). Some landlocked, self-sustaining populations have been created unintentionally from dam construction, such as the upper Roanoke River watershed in North Carolina and Virginia. Others were created intentionally in nontraditional watersheds through stocking programs, such as the Keystone Reservoir population (Arkansas River) in Oklahoma (Combs 1980) and the Lake Texoma population (Red River) in Oklahoma and Texas (Harper and Namminga 1986). Changes in streamflow characteristics can affect spawning activity, egg viability, and postlarval survival (Rulifson and Manooch 1990). Earlier investigations reported that striped bass eggs are slightly heavier than freshwater, with a specific gravity of 1.0003–1.00065 g/cm3, which causes them to sink to the bottom in undisturbed water (Hardy 1978). Slight agitation suspends the eggs within the water column (Mansueti 1958). Unsuspended eggs have a poor chance of survival (Talbot 1966). Nonnative fish stocked in a watershed could conceivably grow and occupy the same habitat as the wild population, only to suffer reproductive failure as adults. This reproductive failure may be caused by poor survival of early life stages, which

FIGURE 1.—Atlantic coast watershed locations of striped bass broodfish collected in 1999 for this study. The upper Roanoke River is also known as the Staunton River at the North Carolina–Virginia border. The Shubenacadie watershed contains the Stewiacke River spawning population.

may not be adapted to the rigors of the nonnatal watershed. If egg density is under genetic control or somehow related to watershed type, we would hypothesize that the heaviest eggs would be produced in higher-physical-energy (e.g., uplanddominated) watersheds, while the lightest eggs would be produced in lower-energy (e.g., coastal) watersheds. If this scenario is true, then reproductive failure could occur by stocking a coastal, tidally influenced watershed with progeny of a population from an upland-dominated watershed and vice versa. To test this hypothesis, we investigated whether differences in egg density and other egg characteristics, such as the surface : volume ratio, relative oil globule size, and lipid content and type, were related to watershed type. Methods Field collection.—We sampled wild-captured broodfish at seven state or private hatcheries to assess nine different striped bass populations rang-

560

BERGEY ET AL.

ing from Georgia (approximately 328N) to Prince Edward Island (approximately 478N; Figure 1). Fish populations from high-physical-energy watersheds included those from two upland-dominated freshwater rivers (Dan River, North Carolina, and the upper Roanoke [Staunton] River, Virginia), an upland tidal-bore river (Stewiacke River, Nova Scotia, a tributary of the Shubenacadie watershed), and an upland tidal river (Miramichi River, New Brunswick). The Miramichi and Stewiacke populations are anadromous, whereas the Dan and upper Roanoke populations are landlocked (since 1950), created from construction of the John H. Kerr Dam and its reservoir on the Roanoke River watershed. Populations representing lower-energy watersheds were from an inland freshwater lake (Lake Lanier, Georgia) and four estuarineinfluenced coastal rivers (Savannah River in Georgia, Pamunkey River in Virginia, and the Choptank and Nanticoke rivers in Maryland). The Lake Lanier population was created by stocking fish of Savannah River origin; the Savannah River population is considered to be endemic and riverine. A subsequent decline of the Savannah population in the 1980s resulted in the use of Lake Lanier fish as a hatchery brood source for restoring the population (Ted Will, Richmond Hill State Fish Hatchery, personal communication). The Pamunkey, Choptank, and Nanticoke rivers all discharge to Chesapeake Bay, and the striped bass populations from these rivers are anadromous. Hatchery personnel collected wild striped bass, which were transported back to hatcheries in aerated collection boxes or small tanks. Protocols for collecting, transporting, holding, and spawning fish were similar among hatchery locations. The Richmond Hill State Fish Hatchery (SFH) in Georgia captured broodfish from both the Savannah River and Lake Lanier. The Watha SFH in North Carolina collected brood from the Dan River, which is a tributary of the Roanoke River. The Dan River population spawns in a region of the river thought to be one of the original historical spawning sites for the Roanoke River population prior to landlocking. The King and Queen SFH in Virginia collected broodfish from both the upper Roanoke (Staunton) River and the Pamunkey River. Manning SFH in Maryland used Nanticoke River broodfish, while the University of Maryland Horn Point Aquaculture Facility collected Choptank River fish. The Nova Scotia Agricultural College collected fish from the Stewiacke River, a tributary of the Shubencadie watershed and the last remaining spawning ground for striped bass in Bay

of Fundy watersheds. However, these fish failed to spawn and therefore wild-caught eggs were used. The Cardigan Fish Hatchery, Prince Edward Island, used Miramichi River broodfish for hatchery operations. All hatcheries collected adult fish by electroshocking, with the exception of the Cardigan facility, which collected by box trap. At the hatcheries, fish were injected with human chorionic gonadotropin above the lateral line between the dorsal fins. All females were injected once, except at the Richmond Hill SFH, which injected two females twice. The Cardigan Fish Hatchery and the King and Queen SFH also injected the males. The anesthetic tricaine methanesulfonate (MS-222) was used only by the Richmond Hill and Cardigan facilities. Tank spawning was the preferred method at all locations except for Richmond Hill SFH, which used the stripspawn method. Once spawned, the eggs were transferred to troughs (King and Queen, Manning, Cardigan), tanks (University of Maryland), or MacDonald jars (Richmond Hill, Watha, and King and Queen), and the adults were returned to the river (Watha), disposed of (King and Queen, Manning), or kept all year (Richmond Hill, University of Maryland, Cardigan). Our experiments were conducted in conjunction with normal hatchery operations and ambient water quality conditions. The number of females examined differed among hatcheries as a direct result of the number of fish successfully induced to spawn. A total of 25 broodfish were used for this study. Wild-caught eggs (Stewiacke only) were also obtained. We used eggs from individual females from Lake Lanier (3 females), Savannah (3), Nanticoke (3), Choptank (4), upper Roanoke (3), and Pamunkey (1) watersheds. The Miramichi River experiments involved three females in one tank, and Dan River fish (n 5 5) were tested with two females per tank in two tanks and one female in a third tank. Strip-spawned eggs were collected directly out of the mixing pan immediately after fertilization, and tank-spawned eggs were collected from the tank with a standard sieve of 60-mm mesh. A 3-mL subsample of unfertilized eggs from each female was placed on wet ice until frozen at 2208C for fatty acid analysis. Fertilized eggs were transferred to MacDonald hatching jars or troughs and subsequently subsampled at periodic intervals with a pipette. Eggs were sampled for diameter and density analysis at four developmental stages (Hardy 1978): stage 1 (,10 h) was defined as fertilized eggs prior to formation of the germ ring; stage 2 (10–18 h) was the period after germ ring


formation but before development of the blastoderm; stage 3 (20–28 h) was the early embryonic phase, after blastoderm formation; and stage 4 ($30 h) was the advanced embryonic phase, characterized by tail release from the embryonic bundle. Stewiacke River broodfish did not spawn in the hatchery, so wild-spawned eggs were collected from the river with a 0.25-m-diameter egg sampling net constructed of 500-mm Nitex mesh. These eggs were transported in a thermos and maintained in an aerated chamber of ambient-temperature river water. Only stages 2–4 were examined from this population. Water quality was analyzed at each of the six hatchery locations and the Stewiacke River. If the hatchery site had separate holding tanks, then each tank was tested. If the hatchery had a centralized water-dispersal system, then the main tank was tested. A Lamotte testing kit was used to measure hardness as CaCO3 (mg/L) and alkalinity as CaCO3 (mg/L). Temperature (8C) was measured with a calibrated mercury thermometer, and pH was measured with an Oakton pH Tester 2 pen. Laboratory analyses.—Fertilized eggs from each female were measured for egg diameter and oil globule size to ascertain whether either parameter changed during embryonic ontogenesis and whether ambient water conditions influenced this change. For each female, approximately 80 eggs were measured at each of the four developmental stages to determine variability in size and oil content. From a subsample of several hundred eggs, half were placed in a petri dish containing ambient hatchery water, and the other half were placed in a petri dish containing USEPA (1987) standardized water, a soft, synthetic freshwater control made with 48 g of sodium bicarbonate, 30 g of calcium sulfate, 30 g of magnesium sulfate, and 2 g of potassium chloride in 1 L of double-distilled, deionized water. The USEPA-standardized water served as the control group. After a 5-min stabilization period, the diameters of 10 randomly selected eggs held in ambient water were measured with an ocular micrometer mounted on a dissecting microscope. Oil globule diameters were measured, and data were standardized to a relative oil globule size as a percentage of the egg diameter. Each egg was used only once and then discarded. This procedure was repeated for 10 eggs held in control water for each developmental stage. We tested for the presence of ontogenetic shifts in egg density of fertilized eggs from each female, and we determined whether ambient water con-

561

FIGURE 2.—Experimental design to test striped bass egg densities of individuals from nine Atlantic coast populations under ambient hatchery and USEPA (1987) control water quality conditions at four stages of development and 12 salinities.

ditions influenced this density shift. Egg density was determined by placing 12 groups of randomly selected eggs (10 per group) held in ambient water into salinity solutions ranging from 0‰ to 11‰ and standardized to 178C (Rulifson and Tull 1999). Approximately 1,000 eggs from each female were used for these analyses. Each group was gently placed in a test cylinder with an eyedropper, and the number of eggs sinking, floating, or remaining neutral (no longer moving down the cylinder) was recorded. This procedure was repeated for 12 groups of 10 randomly selected eggs held in USEPA control water (Figure 2). Eggs were tested only once and then discarded. The experiment was repeated with new groups of eggs at the three remaining developmental stages. Lipids were wet extracted from stage-1 fertilized egg samples with a chloroform–methanol–water (2:2:1) solution according to the method of Bligh and Dyer (1959). Neutral lipids and phospholipids were separated with a Supelco LC-Si solid phase extraction tube. Neutral lipids are those stored in each egg for embryonic development. Phospholipids are those present in all cell membranes of the adult fish, including membranes of developing eggs. Neutral lipids were eluted with 5% methanol in chloroform, and phospholipids were eluted with pure methanol (Gallagher et al. 1998). Neutral lipids and phospholipids were methylated with toluene and methalonic base reagent (Gallagher et al. 1998). Methylated samples were dissolved in hexane for subsequent fatty acid analysis with a Varian 3740 gas chromatograph equipped with a flame ionization detector and an automated integrator recorder (Gallagher et al. 1989). Individual fatty acids were identified by comparing the retention time of the fatty acid to the retention time of known

562

BERGEY ET AL.

TABLE 1.—Water quality from experiments conducted with nine Atlantic coast striped bass populations during spring 1999 at six hatcheries and the Stewiacke River (SFH 5 State Fish Hatchery). Hatchery/collection site Cardigan Fish Hatchery, Prince Edward Island Stewiacke River, Nova Scotia (wildcollected) University of Maryland, Horn Point Aquaculture Facility Manning SFH, Maryland King and Queen SFH, Virginia Watha SFH, North Carolina Richmond Hill SFH, Georgia

Hardness (mg/L)

Alkalinity (mg/L)

Temperature (8C)

pH

Miramichi River

84–88

90–104

12.5–16.0

7.6–7.7

Stewiacke River

138

50

19.4–19.9

7.7

Choptank River

160–420

300–344

17.5–19.0

8.2–8.3

Nanticoke River Pamunkey River Upper Roanoke (Staunton) River Dan River Savannah River Lake Lanier

104–120 301–304 40–72 200–212 90 92

129–154 72–74 54–70 228–240 118 132

18.5 20.0–20.5 19.0–20.0 18.0 19.0 19.0

7.8–8.2 7.3–7.4 7.5–7.6 8.0 7.8 8.0

Origin of broodfish

fatty acid methyl ester standards, or by log plots of retention times (McNair and Bonelli 1968). Data analyses.—Egg diameter, oil content, and egg density were compared among groups by analysis of variance (ANOVA) with Duncan’s grouping criteria for tabular presentation (SAS Institute 1999). Eggs from individual fish were tested for salinity responses at each of the four developmental stages held under ambient and control water conditions. Data were transformed as loge 1 1 prior to analysis, and the significance level was set at 0.05. A cluster analysis (SAS Institute 1999) was used to determine which watersheds were similar in striped bass egg characteristics based on the variables of egg diameter, oil globule diameter, and egg density. The surface : volume (S:V) ratio of the mean egg diameter for a population was calculated from the following geometric formulae for the surface (S) and volume (V) of a sphere: S 5 4pr 2 ,

and

V 5 (4/3)pr 3 ,

where r is the radius. A factor analysis (SAS Institute 1999) was used to determine whether the complement of neutral fatty acids stored in the eggs was different among populations but similar in composition by watershed type. Fatty acid data were arcsine-square-root transformed and combined with the egg diameter and oil globule size variables for analysis. To facilitate statistical analysis, a value of 0.01 was used for fatty acids that were not detected but that were present in one or more populations tested. Cluster analysis was used to group and visually depict the relationships. Results Water Quality Hardness and alkalinity varied among and within sampling locations (Table 1). Four of the seven

locations had hardness and alkalinity values above 100 mg/L CaCO3. Facilities with hardness lower than 100 mg/L included Richmond Hill SFH, King and Queen SFH, and Cardigan Fish Hatchery. Eggs reared in water with hardness lower than 80 mg/L had poor survival, usually dying before hatching. King and Queen SFH, which had alkalinity values between 54 and 74 mg/L and hardness values between 40 and 72 mg/L, had the lowest larval survival of the study. Water temperature remained constant within facilities but fluctuated among locations. Temperature ranged between a high of 20.58C at King and Queen SFH and a low of 12.58C at Cardigan Fish Hatchery, but most temperatures were between 18.08C and 20.58C (Table 1). The pH ranged from 7.3 to 8.3 among locations. Within individual hatcheries, pH did not vary by more than 0.4 units. Egg Diameter Water hardening of striped bass eggs occurred within 2.5 h postfertilization, depending on water hardness. Mean egg diameter did not differ significantly between ambient water-hardened eggs and those hardened in USEPA-standardized water (F 5 0.23; df 5 1, 1,538; P 5 0.63). Diameter of water-hardened eggs ranged between 1.5 and 4.2 mm and varied significantly among populations (F 5 12.09; df 5 8, 12; P 5 0.0001). The largest average (6SD) egg diameter (3.67 6 0.10 mm) was that of the Stewiacke population (Table 2). Mean egg diameter ranged from 2.57 to 2.92 mm for the Miramichi, Savannah, and upper Roanoke populations. Mean egg diameter of the Lake Lanier population (2.35 6 0.27 mm) was smaller but not significantly different from the Savannah parental stock (2.77 6 0.19 mm) (Table 2). Smallest eggs were from the Choptank (2.10 mm), Nanticoke


(2.02 mm), Dan (2.00 mm), and Pamunkey (1.84 mm) populations (Figure 3). Mean egg diameter was not related to the minimum ambient water hardness measured at the hatcheries (F 5 2.36; df 5 1, 8; P 5 0.17). Egg diameter differed significantly among watersheds at each embryonic stage of development (Table 3). In addition, several populations exhibited significant changes in egg diameter during embryonic ontogenesis. The smallest egg diameters in the Miramichi and Lake Lanier populations were observed at stage 4, but in the Savannah population, this stage had the largest diameter. Prespawn weight of females (5–11 kg) was not correlated to diameter of eggs produced. The S:V ratios of eggs from higher-energy watersheds were lower than those of lower-energy watersheds, and ranked as follows (in ascending order): Stewiacke River (1.6:1), upper Roanoke River (2.0:1), Savannah River (2.2:1), Miramichi River (2.3:1), Lake Lanier (2.6:1), Choptank River (2.8:1), Nanticoke River (3.0:1), Dan River (3.0: 1), and Pamunkey River (3.3:1). Oil Globules Mean oil globule size ranged from 0.57 to 0.85 mm in diameter and differed significantly among watersheds (Table 2). The largest relative oil globule size (46.5%) was found in the Pamunkey population, which was not statistically different from the Nanticoke population (39.3%). The upper Roanoke population had the smallest relative oil globule size (19.6%). Other populations from highenergy watersheds exhibiting similarly small relative oil globule sizes included the Miramichi (26.8%) and the Stewiacke (22.7%) rivers (Figure 3). Mean relative oil globule size was significantly related to nominal values of ambient hatchery water hardness (F 5 8.24; df 5 1, 8; P 5 0.02; r2 5 0.54). Relative oil globule size differed significantly among populations at each developmental stage (Table 4). Six of nine populations showed significant changes in relative oil globule size with embryonic ontogenesis. Four populations (Nanticoke, upper Roanoke, Savannah, and Lake Lanier) showed a decrease in relative oil globule size from stage 1 to stage 4. Only the Miramichi population showed an increase in relative oil globule size, a direct effect of reduced egg diameter at stage 4. Egg Density Egg densities differed significantly among populations, with the heaviest eggs (highest specific gravity) produced by females collected from highenergy watersheds (Table 2). The Miramichi and

563

upper Roanoke eggs showed no significant changes in the percentage of eggs sinking or floating over the range of salinities used in the experiment, which equated to egg densities of over 1.0087 g/ cm3. Eggs from the Dan River population exhibited a similar trend, with the exception that stage1 and stage-2 eggs showed significant salinity responses at 10–11‰ (F 5 4.33 and 4.35; df 5 11, 61; P , 0.0001), which was not observed at stages 3 and 4. The Choptank, Nanticoke, Pamunkey, and Savannah populations showed significant shifts in salinity response for all developmental stages beginning at 1–2‰, and no significant changes after 3‰. Eggs from Lake Lanier females exhibited similar responses to the salinity gradient at stages 1 and 4, but at stages 2 and 3, some eggs continued to sink through the range of salinities tested, suggesting a slight increase in egg density during embryonic ontogenesis. Eggs from the Stewiacke watershed did not respond to salinity as expected, with no significant shifts to the salinity gradient after 2‰. Egg density was not significantly related to nominal values of ambient hatchery water hardness (F 5 0.55; df 5 1, 8; P 5 0.48). Egg diameter, oil globule diameter, and egg density showed two distinct groupings of populations (Figure 4). Populations from higher-energy watersheds (Dan, Miramichi, Roanoke, and Stewiacke) grouped separately from populations from lower-energy watersheds (Choptank, Nanticoke, Lake Lanier, Savannah, and Pamunkey). Fatty Acids Neutral lipids.—Fatty acid distributions of oil globule neutral lipids differed significantly among populations, correlating with the relative physical energy of the watersheds. Monounsaturated fatty acids dominated (51.9–75.9%) in eggs from higher-energy watersheds (Dan, upper Roanoke, Miramichi, and Stewiacke), and polyunsaturated fatty acids dominated (59.1–73.4%) in eggs from low-energy watersheds (Lake Lanier, Pamunkey, Choptank, and Nanticoke; Table 2). Eggs from Savannah River fish were more similar to those from Lake Lanier than to populations from higherenergy watersheds. The monounsaturated fatty acid 18:1(n-7)/(n-9) made up most (37.6–51.0%) of the total monounsaturated fatty acids present in eggs from high-energy watersheds, followed by 20:1(n-9). (In this notation, the number to the left of the colon is the number of carbon atoms, the number immediately to the right of the colon is the number of double bonds, and the number in parentheses indicates the position of the first dou-

564

BERGEY ET AL.

TABLE 2.—Summary of egg characteristics determined for nine Atlantic coast populations of striped bass during spring 1999. Means with the same letter (Duncan grouping) are not significantly different (P . 0.05) within the column. Degrees of freedom and P-values are as follows: egg diameter (df 5 8, 12; P , 0.0001), relative oil globule size (df 5 8, 12; P 5 0.0003), neutral lipids (df 5 8, 11; saturated [Sat] P 5 0.3185; monounsaturated [Mono] P , 0.0001; polyunsaturated [Poly] P 5 0.0005), and phospholipids (df 5 8, 11; polyunsaturated P 5 0.0005).

Population Miramichi Stewiacke Choptank Nanticoke Pamunkey Upper Roanoke Dan Savannah Lake Lanier

Egg diameter (mm; 6SD) 2.57 3.67 2.10 2.02 1.84 2.92 2.00 2.77 2.35

6 6 6 6 6 6 6 6 6

0.23 0.10 0.20 0.08 0.01 0.10 0.15 0.19 0.27

yx z xwv wv v y wv y yxw

Oil diameter (mm; 6SD) 0.68 0.83 0.75 0.79 0.85 0.57 0.62 0.69 0.76

6 6 6 6 6 6 6 6 6

0.01 0.02 0.03 0.03 0.01 0.02 0.06 0.05 0.20

yxw zy zyx zy z w xw yxw zy

ble bond from the methyl end.) No such pattern was observed in eggs from low-energy watersheds, which contained much less total monounsaturated lipid and much smaller amounts of 18:1 and 20:1 (Table 5). Monounsaturated fatty acids in eggs from low-energy watersheds had more of the shorter chain fatty acids, 15:1 and 16:1. Eggs from the Pamunkey system were unusual in that they had very low levels of monounsaturated fatty acids and high levels of polyunsaturated fatty acids (Table 5). Eggs from low-energy watersheds were dominated by polyunsaturated fatty acids of 18 carbons or less. Eggs from the Savannah, Lake Lanier, and Choptank watersheds had large amounts of 16:3(n3) and/or 16:4(n-3)/(n-4). The Choptank, along with the Pamunkey and Nanticoke, also had high levels of 16:2(n-4) and/or 16:3(n-4). Savannah eggs also had measurable amounts of 20:5(n-3) ecosapentanoic acid (EPA) and 22:6(n-3) decosahexanoic acid (DHA). Lake Lanier eggs had only low or immeasurable levels of these fatty acids, but contained a large amount of 18:3(n-3), as did eggs from the Pamunkey and Nanticoke rivers. The Choptank and Nanticoke eggs also contained high levels of 18:4(n-3). The only polyunsaturated fatty acid found in measurable amounts in the neutral lipids of Miramichi and Stewiacke eggs was DHA, and 18:2(n-6) was found in Miramichi eggs only. Fatty acid composition of the neutral lipid fraction showed two distinct groupings (Figure 5). The Dan and Savannah populations were reversed, but otherwise the groupings were the same as for the relationships based on egg diameter, oil globule diameter, and density (Figure 4). Phospholipids.—The total saturated fatty acid content of egg membranes from all systems was

Relative oil globule size (%; 6SD) 26.8 22.7 36.4 39.3 46.5 19.6 31.7 25.1 32.8

6 6 6 6 6 6 6 6 6

2.611 xwv 0.40 wv 2.70 y 2.66 zy 0.75 z 0.88 v 5.37 yxw 1.70 xwv 9.02 yx

Salinity at .80% floating (‰)

Specific gravity (g/cm 3 )

.11 2.5 3.5 3.0 2.5 .11 .11 1.5 4.5

$1.0087 1.0018 1.0025 1.0022 1.0018 $1.0087 $1.0087 1.0010 1.0033

variable, ranging from a high of 38.9% (Lake Lanier) to a low of 6.3% (Pamunkey; Table 2). The dominant saturated fatty acid in the phospholipid fraction was 16:0, except in eggs from the Dan and Pamunkey watersheds, which contained much smaller amounts of saturated fatty acids than the other populations and were dominated by 20:0 and 18:0, respectively (Table 6). The fatty acid 18:0 also occurred in relatively large amounts in eggs from the other systems. Only eggs from the Dan River population had significantly higher levels of total monounsaturated fatty acids than the other systems (Table 2). Likewise, the monounsaturated fatty acid 18:1(n9)/(n-7) was the dominant monounsaturated fatty acid found in eggs from the Savannah, Lake Lanier, and upper Roanoke systems, whereas the Choptank and Nanticoke rivers had similar amounts of 18:1(n-9)/(n-7) and 16:1(n-7) (Table 6). No measurable amount of any monounsaturated fatty acid was found in egg phospholipids from fish collected from the Pamunkey watershed. In terms of polyunsaturated fatty acids, the eggs from the Pamunkey and Savannah watersheds had significantly higher levels compared to the Dan River eggs (Table 2). Contrary to the results from neutral lipid fatty acid content, longchain fatty acids ($ 20 carbons) dominated the polyunsaturated fatty acids of the phospholipid fraction of eggs from all watersheds. Eggs from both the Savannah River and Lake Lanier systems contained 20:4(n-6) arachidonic acid (AA), 20: 5(n-3) EPA, and 22:6(n-3) DHA in increasing amounts. The Choptank and Nanticoke eggs also contained these fatty acids but in smaller amounts (Table 6). Fatty acid composition of the phos-

565


TABLE 2.—Extended.

Mean fatty acid content (%) Phospholipids

Neutral lipids Population

Sat

Miramichi Stewiacke Choptank Nanticoke Pamunkey Upper Roanoke Dan Savannah Lake Lanier

17.1 6.0 10.4 10.1 11.6 12.0 14.3 5.7 1.6

Mono 64.9 75.9 12.0 6.2 0.8 51.9 61.4 15.8 16.5

y z wv vu u x y w w

pholipid fraction showed that none of the populations were closely related in fatty acid composition in the membrane structure (Figure 6). The Choptank and Nanticoke populations grouped together, and the Lake Lanier and Savannah populations were similar. The Pamunkey and Dan populations were the least similar to the other populations in phospholipid fatty acid composition. Discussion The number and size of fish used in this study were a direct result of local hatchery operations, including successful capture of wild broodstock, health maintenance in the hatchery during the prespawn holding period, and successful inducement of spawning. We accepted the assumption by hatchery personnel that broodfish originated from the watershed of collection. The exception was broodfish from Lake Lanier, which was stocked with progeny from the Savannah River population several decades previously (T. Reinhart, University of Georgia, Athens, personal communication). In some cases, the number of experiments performed here represented the entire hatchery production; the progeny of one fish may have represented the entire variability of the population to be stocked. Additional experiments on more individual fish from the Pamunkey and Stewiacke watersheds is encouraged to confirm trends found in this study. Water hardness of less than 80 mg/L had a negative effect on larval survival, as would be expected from previous studies (Hoar and Randall 1988), but other water quality parameters were maintained at as close to optimal conditions (Har-

Poly 7.6 3.6 69.8 59.8 73.4 18.0 6.4 60.8 59.1

y y z z z y y z z

Sat

Mono

28.8 28.9 6.3 36.8 12.1 29.8 38.9

13.1 12.6 0.0 17.9 47.5 20.8 14.2

Poly

36.6 40.1 53.5 36.6 16.5 55.5 35.8

zy zy z zy y z zy

rell 1997) as is possible in hatcheries. Water temperatures between 16.58C and 208C are considered optimal for egg and larval development in hatcheries. Ideally, pH should be around 7.4 (Harrell et al. 1990), but the favorable pH range for striped bass larvae and young is reported at between 6.0 and 10.0 (Regan et al. 1968). Larvae are extremely sensitive to a sharp change of pH even within these limits, but no sudden changes in pH were observed during the study. Although Harrell et al. (1990) reported that egg diameter ranges from approximately 1.2 to 4.2 mm depending on hardness and salinity during water absorption through the chorion, there was no correlation between egg diameter and the water hardening of eggs in ambient water versus standardized water. Egg size varied significantly among watersheds, but the cause and importance are unknown. Chambers and Leggett (1996) demonstrated maternal influences, including size of the mother, on variation in egg size. Maternal influence was not apparent in this study when weight of the female broodfish was plotted against egg diameter, probably due to the narrow range of maternal weights (5–11 kg). Variation in egg size can also arise from effects of stress and other environmental factors, including poor feeding or diet, and adverse temperature, light, and water quality (Hoar and Randall 1988). In general, striped bass eggs from lower-energy systems had greater S:V ratios, an aspect separate from buoyancy that is likely important to maintaining position within the water column during embryonic development. A high S:V ratio translates to increased friction between the egg surface and supporting water mass, which creates drag and slows the rate of egg descent through the water

566

BERGEY ET AL.

FIGURE 3.—Mean (6SE) values of striped bass egg diameter (mm), relative oil globule size (%), and egg density (g/cm3) for nine Atlantic coast populations.

column. This phenomenon is commonly seen in zooplankton, which reduce their sinking rate by increasing drag through surface water friction (Castro and Huber 2002).

Oil globule size and egg density were correlated with relative energy of the watershed. Mansueti (1958) reported that the average oil globule diameter of striped bass eggs was 0.61 mm (range 0.40–0.85 mm), and the oil globule size remained the same during early developmental stages. Except for one sample of Lake Lanier eggs, and the eggs from the Miramichi system, our results support his findings in the actual diameter measurements of both egg and oil globule. However, when the oil globule diameter was standardized to the overall egg diameter, relative oil globule size differed significantly with respect to specific watersheds, thereby suggesting that oil globules in eggs from different watersheds have different characteristics. Eggs from higher-energy systems had smaller relative oil globule size and heavier (denser) eggs. For example, eggs from the Dan, upper Roanoke, and Miramichi populations had low relative oil globule size and high egg density. The exception was the Stewiacke system, for which relative oil globule size was small and test results suggested low-density eggs. However, the Stewiacke sample was the only wild-caught egg sample tested, and the sample size was small. The Stewiacke results are contradicted by the results of Rulifson and Tull (1999), which indicated highdensity eggs based on a much larger sample size. Results of our study clearly show that egg densities and subsequent buoyancy characteristics are different among populations. Striped bass eggs have been reported as buoyant or semibuoyant, being found at various levels within the water column and floating easily with little agitation (e.g., Mansueti 1958; Hardy 1978). Broodfish believed to have sinking (or more dense) eggs by hatchery managers were collected from floodplain nontidal or upland tidal watersheds. Fish believed to have floating (or less dense) eggs were collected from low-physical-energy watersheds, primarily those discharging into Chesapeake Bay. Buoyancy characteristics are critical to egg survival by giving the egg the ability to avoid sinking to the bottom and suffocating, or floating to the surface and thereby becoming stranded on the banks and dying from desiccation. Rulifson and Tull (1999) reported mass strandings of striped bass eggs on Stewiacke River mudflats at low tide during peak spawning activity. Whether those eggs remained moist and viable until resuspended by the flood tide was unknown. Based on the results of our study, a similar problem could be created if fish from populations with less dense eggs (e.g., Chesapeake Bay origin) were used for stock enhancement of an upland

567


TABLE 3.—Results from analysis of variance for egg diameter during embryonic ontogenesis and differences at each developmental stage for nine Atlantic coast populations of striped bass (stage 1 5 ,10 h; stage 2 5 10–18 h; stage 3 5 20–28 h; stage 4 5 $30 h) Mean values (log e 1 1) with the same lowercase letters (Duncan grouping) are not significantly different between stages; means with the same uppercase letters are not significantly different between populations. Stage of development Population Miramichi Stewiacke Choptank Nanticoke Pamunkey Upper Roanoke Dan Savannah Lake Lanier All populations df F P

df 3, 2, 3, 3, 3, 3, 3, 3, 3,

76 57 316 236 76 136 236 176 196

F

P

1

24.37 2.10 1.08 1.20 0.23 0.03 2.88 2.98 12.32

,0.0001 0.1314 0.3591 0.3104 0.8737 0.9939 0.0366 0.0327 ,0.0001

1.02 Yz 0.75 0.70 0.60 1.07 0.71 0.98 0.83

2

W W V Z Wz Yy Xy

7, 412 85.18 ,0.0001

floodplain river population, which would normally have heavier and less buoyant eggs. If buoyancy is under genetic control, theoretically the supplanted fish would grow and utilize native fish habitat, then produce eggs with characteristics incompatible with the watershed. The result would be a disruption in the life cycle, ending with no second generation. Results of our study indicated that fatty acid composition in eggs was similar across the coastal range of striped bass for watersheds of similar physical and energy characteristics. Lipids are sources of metabolic energy during female gonad development and are essential materials for formation of cell and tissue membranes (Sargent 1995). Fatty acid and total lipid content of fish differ greatly between species and within a species

1.00 Xz 1.32 Z 0.73 V 0.69 V 0.61 U 1.07 Y 0.69 Vzy 1.01 Xzy 0.92 Wz 8, 391 166.38 ,0.0001

3

4

0.93 1.31 0.77 0.71 0.61 1.07 0.67 1.01 0.83

Wy Z U T S Y Ty Xzy Vy

8, 351 154.43 ,0.0001

0.82 1.27 0.77 0.70 0.61 1.07 0.69 1.06 0.77

Xx Z X W V Y Wzy Yz Xx

8, 351 130.95 ,0.0001

due to abiotic factors, diet, developmental stage, sex, or seasonal differences (Gallagher et al. 1989). Total lipid accumulation depends upon the availability and type of food (Shulman and Love 1999). Worthington and Lovell (1973) examined channel catfish Ictalurus punctatus and determined that genetic and other factors accounted for about 5% of the variance in total fatty acid composition of body lipids, whereas diet accounted for 95%. Harrell and Woods (1995) and Seaborn et al. (2000) distinguished cultured striped bass from wild fish by differences in fatty acid content due to dietary differences. Population differentiation based on fatty acids was reported by Grahl-Nielsen and Mjaavatten (1992), who identified striped bass spawning stocks among river systems by the fatty acid content of heart tissue lipid (mainly phos-

TABLE 4.—Results from analysis of variance for relative oil globule size (% of egg diameter) during embryonic ontogenesis and differences at each developmental stage for nine Atlantic coast populations of striped bass. Mean values are loge + 1. See Table 3 for additional details. Stage of development Population Miramichi Stewiacke Choptank Nanticoke Pamunkey Upper Roanoke Dan Savannah Lake Lanier All populations df F P

df 3, 2, 3, 3, 3, 3, 3, 3, 3,

76 57 316 236 76 136 236 176 196

F

P

26.4 0.97 0.16 8.88 4.76 6.24 1.82 4.28 42.08

0.0001 0.3861 0.9200 0.0001 0.0043 0.0005 0.1400 0.0060 0.0001

1

2

3.19 Vx 3.56 3.71 3.85 3.00 3.41 3.25 3.69

X Yz Zz Uz W Vz Yz

7, 412 162.05 ,0.0001

3.23 3.09 3.56 3.64 3.81 2.95 3.41 3.21 3.46 8, 391 108.29 ,0.0001

3 Vx U X Yx Zy Ty W Vzy Wy

3.31 3.13 3.56 3.66 3.85 2.95 3.49 3.21 3.27

4 Vy T X Yyx Zz Sy W Uzy VUx

8, 351 136.42 ,0.0001

3.40 3.12 3.57 3.67 3.84 2.95 3.42 3.16 3.28

Wz U X Yy Zz Ty W Uy Vx

8, 351 139.22 ,0.0001

568

BERGEY ET AL.

FIGURE 4.—Groupings of Atlantic coast striped bass populations by similarities in egg diameter, oil globule diameter, and egg density.

TABLE 5.—Mean percent fatty acid content of neutral lipids from striped bass eggs collected from nine Atlantic coast populations during spring 1999 (n 5 number of females used in each experiment). See text for a description of fatty acid notation.

Fatty acid

Miramichi Stewiacke Choptank River River (wild- River (n 5 1) collected) (n 5 4)

Saturated 14:0 16:0 18:0 20:0 Monounsaturated 15:1 16:1 (n-7) 18:1 (n-9)/(n-7) 20:1 (n-9) Polyunsaturated 16:2 (n-4) 16:3 (n-4) 16:3 (n-3) 16:4 (n-4)/(n-3) 18:2 (n-6) 18:3 (n-4) 18:3 (n-3) 18:4 (n-6) 18:4 (n-3) 20:4 (n-6) 20:4 (n-3) 20:5 (n-3) 22:6 (n-3) Total fatty acids in neutral Saturated Monounsaturated Polyunsaturated

Nanticoke Pamunkey River River (n 5 3) (n 5 1) 2.6 1.6 5.9

8.2

0.3 2.9 7.1 0.1

0.4 50.1 14.3

5.5 3.0 3.4 0.2

4.0 1.1 1.0

0.8

0.4 28.7

12.6 19.3

34.7 9.6

15.2 0.5 0.3 5.7

2.7 0.4

8.9

6.0

51.0 24.9

4.6

2.9

3.4

lipid fraction 17.1 6.0 64.9 75.9 7.4 3.4

0.3 3.6 7.7

Upper Roanoke River (n 5 3)

Dan River (n 5 3)

Savannah River (n 5 3)

Lake Lanier (n 5 1)

0.4 3.1

0.1 6.1

0.8 1.8 3.1

0.3 1.3

8.4

8.1

0.9 2.1 37.6 11.3

0.7 44.8 15.9

5.0 5.4 5.4 0.1

5.8 3.4 6.7 0.6

3.8

0.9 0.6 14.2 22.0 0.3

37.0 8.1 1.1

6.7

11.2

0.2 4.6 20.1

16.0 0.3 1.8 0.4 0.3

11.1 0.6 1.5

0.1 2.3 0.2

5.2 0.1 0.4 0.9 1.1

10.4 12.0 69.8

10.1 6.2 59.8

11.6 0.8 73.4

3.2

12.0 51.9 17.9

1.2 1.3

14.3 61.4 6.3

1.1 6.1 8.0 0.5 0.4 3.2 2.6

7.7 3.0

5.7 15.8 60.7

1.6 16.5 59.0

0.8

0.8


569

FIGURE 5.—Groupings of Atlantic coast striped bass populations by similarity in the fatty acids of neutral lipids listed in Table 5. Fatty acids not detected in some populations were assigned a value of 0.01 (the lowest detectable level) to facilitate analysis.

pholipids). The major saturated (16:0 and 18:0) and monounsaturated (18:1(n-9)/(n-7)) fatty acids found in this study are common for fish (Harrell and Woods 1995; Gallagher et al. 1998). The dominance of 16:0 and 18:0 in the neutral lipids indicates that, as with other fish (Rainuzzo et al. 1997), these fatty acids are primary sources of stored energy, because the neutral fraction is composed mostly of storage lipids. However, fatty acid composition differed among populations from lower-energy watersheds compared to higherenergy systems. Specifically, 16:0 and 18:1(n-9)/ (n-7) occurred in higher amounts in higher-energy systems, while 18:0 occurred in higher amounts in lower-energy watersheds. Most striking was the presence of high amounts of polyunsaturated fatty acids as short-chain omega-3 and omega-4 fatty acids, such as 16:3(n-3), 16:3(n-4), and 16:4(n-4)/ (n-3), in the neutral lipids of floating eggs. Because the type of fatty acid should not change buoyancy (Gee et al. 1974), the reason for the incorporation of these fatty acids into neutral lipids is not clear but may be related to diet in these systems, and is consistent with a freshwater forage base. As expected, long-chain omega-3 fatty acids occurred in small amounts in the neutral fatty acids. These fatty acids are important to membrane structure

and are found primarily in the phospholipid fraction. In this study, there were few significant differences in the total saturated, monounsaturated, and polyunsaturated fatty acid content of the phospholipids of eggs. Such results would be expected since the phospholipid content is associated with essential membrane structures. However, the phospholipid fractions from floating eggs of lowerenergy watersheds contained significantly more 20:4(n-6), 20:5(n-3), and 22:6(n-3) than sinking eggs from higher-energy systems, the Dan and upper Roanoke watersheds. These findings are consistent with those of Gallagher et al. (1998), who reported lower levels of these fatty acids in sinking eggs from higher-energy systems. Our data are also consistent with expected levels in upland freshwater rivers such as the Dan and upper Roanoke. The neutral and phospholipid fraction of eggs from Pamunkey River striped bass were unrelated to those of all other populations examined in this study, or to those fatty acids reported in other studies. However, additional sampling and increased sample size are necessary to confirm such differences. Low levels of long-chain omega-3 fatty acids are of concern because larvae of many marine fishes require highly unsaturated fatty acids of the

570

BERGEY ET AL.

TABLE 6.—Mean percent fatty acid content of phospholipids from striped bass eggs collected from seven Atlantic coast populations during spring 1999 (n 5 number of females used in each experiment). Miramichi and Stewiacke populations were not tested.

Fatty acid

Choptank River (n 5 4)

Saturated 14:0 1.3 16:0 12.6 18:0 11.7 20:0 3.2 Monounsaturated 15:1 1.5 16:1 (n-7) 5.8 18:1 (n-9)/(n-7) 5.0 20:1 (n-9) 0.8 Polyunsaturated 16:2 (n-4) 1.7 16:3 (n-4) 1.4 18:2 (n-6) 0.2 18:3 (n-4) 0.4 18:3 (n-3) 2.8 18:4 (n-3) 20:2 (n-9) 20:4 (n-6) 1.4 20:4 (n-3) 6.2 20:5 (n-3) 5.4 22:5 (n-3) 1.1 22:6 (n-3) 4.9 23:2 (n-4) Total fatty acids in phospholipid fraction Saturated 28.8 Monounsaturated 13.1 Polyunsaturated 25.4

Nanticoke River (n 5 3) 3.4 17.3 8.3

Pamunkey River (n 5 1)

Upper Roanoke River (n 5 3)

Dan River (n 5 2)

Savannah River (n 5 1)

Lake Lanier (n 5 1)

3.0

6.3

20.3 16.6

1.1 16.5 12.2

0.6 27.2 11.1

2.0 18.8

1.1 3.3 9.9

9.1 0.2 5.4 6.4 0.7 4.6 6.8

1.9 16.0

6.4

10.2 30.3 7.1

3.9 4.3

1.0 0.2 8.0

1.5 1.4

1.6 4.3

3.8

6.2 6.6 11.2 5.9 8.9

15.0 1.4 4.3 4.8

12.9

28.9 12.6 45.0

6.3 0.0 53.4

4.9 12.6 9.9

36.8 17.9 36.9

3.6

12.1 47.5 16.5

6.2 0.3 15.1 3.5 29.1

10.0

29.8 20.8 55.4

38.9 14.2 35.7

8.9 14.0

FIGURE 6.—Groupings of Atlantic coast striped bass populations by similarity in the fatty acids of phospholipids listed in Table 6. Fatty acids not detected in some populations were assigned a value of 0.01 (the lowest detectable level) to facilitate analysis.

571


n-3 series (Rainuzzo et al. 1997). Lipids affect the egg quality of many fish species. A deficiency of n-3 unsaturated fatty acids in broodstock can negatively affect fecundity, fertilization rate, and hatching rate (Rainuzzo et al. 1997). However, during our study, hatchery personnel reported no acute hatching declines. Whether these differences in lipids are the result of adult female striped bass foraging on different food webs in the different watersheds is unclear. Factors other than food may be important in determining lipid content and fatty acid profiles of the eggs. For example, Rainuzzo et al. (1997) found that lipid levels and fatty acid content in turbot Scophthalmus maximus can vary for different egg batches from the same female or among females even when the diet is constant. However, our results clearly showed that fatty acid variability among individuals was small enough to allow differentiation of populations. It is interesting to note that of the higher-physicalenergy watersheds, the Dan and upper Roanoke provide habitat for landlocked upland populations, and the coastal Miramichi and Stewiacke contain anadromous populations. For lower-energy watersheds, the three Chesapeake Bay rivers have anadromous populations, whereas Lake Lanier is a landlocked population. Because ocean food webs are different from riverine food webs, one could conclude that other factors along with food web (e.g., genetics) influence fatty acid composition of developing striped bass eggs. Our results indicate that egg characteristics, including egg diameter, egg density, relative oil globule size, total lipid content, and fatty acid content, are correlated with watershed type and that this relationship should be considered when choosing broodstock for population enhancement and restoration programs. If a hatchery is to produce fish for stocking in natural bodies of water to initiate or supplement natural stocking, the obvious choice is to collect the fish from the existing population in the body of water designated to receive the stocking. If the objective of stock restoration or enhancement is to restore and increase the number of naturally reproducing striped bass, broodfish originating from a similar watershed should be selected to ensure that early life history stages will have the best chance of survival to complete the life cycle. Because watershed type and natal striped bass populations are compatible, water flow characteristics in a watershed should only be manipulated after careful consideration of spawning habitats and the striped bass early life history. Anthropogenic changes to striped bass watersheds

are numerous, including the Savannah River tide gate, the Santee–Cooper rediversion project, and hydropower production on the lower Roanoke River. Continued demands for navigation by channelization, electrical power production, and water diversion through interbasin transfer can and will substantially alter watershed flow patterns, especially if a watershed has cumulative impacts from multiple projects. Acknowledgments This project would not have been possible without the help of several state, provincial, and private organizations and their personnel: Richmond Hill SFH, Georgia (T. Meronick, T. Reinert); Watha SFH, North Carolina (J. Evans); King and Queen SFH, Virginia (C. Dahlem); Joseph Manning SFH, Maryland (B. Florence, J. Van Tasslee); University of Maryland Horn Point Aquaculture Facility (R. Harrell); Nova Scotia Agricultural College (J. Duston, P. MacIsaac); and Cardigan Fish Hatchery, Prince Edward Island (R. Angus). We thank D. Holbert of East Carolina University for statistical assistance. We also thank T. Reinert, E. Zlokovitz, and an anonymous reviewer for valuable comments on improving the manuscript. The first author funded this project with financial assistance from the Department of Biology, East Carolina University, and a grant to M. L. Gallagher. References Bligh, E. G., and W. J. Dyer. 1959. A rapid method of total lipid extraction and purification. Canadian Journal of Biochemistry and Physiology 37:911– 917. Castro, P., and M. E. Huber. 2002. Marine biology, 4th edition. McGraw-Hill, New York. Chambers, R. C., and W. C. Leggett. 1996. Maternal influences on variation of egg sizes in temperate marine fishes. American Zoologist 36:180–196. Combs, D. L. 1980. Striped bass spawning in the Arkansas River tributary of Keystone Reservoir, Oklahoma. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 33(1979):371–383. Forrester, C. R., A. E. Peden, and R. M. Wilson. 1972. First records of striped bass, (Morone saxatilis), in British Columbia waters. Journal of the Fisheries Board of Canada 29:337–339. Gallagher, M. L., S. H. McLeod, and R. A. Rulifson. 1989. Seasonal variations in fatty acids of striped bass, Morone saxatilis. Journal of the World Aquaculture Society 20:38–45. Gallagher, M. L., L. Paramore, D. Alves, and R. A. Rulifson. 1998. Comparison of phospholipid fatty acid composition of wild and cultured striped bass eggs. Journal of Fish Biology 52:1218–1228.

572

BERGEY ET AL.

Gee, J. H., K. Machniak, and S. M. Chalanchuk. 1974. Adjustment of buoyancy and excess internal pressure of swim bladder gases in some North American freshwater species. Journal of the Fisheries Board of Canada 31:1139–1141. Grahl-Nielsen, O., and O. Mjaavatten. 1992. Discrimination of striped bass stocks: a new method based on chemometry of the fatty acid profile in heart tissue. Transactions of the American Fisheries Society 121:307–314. Hardy, J. D., Jr., editor. 1978. Pages 91–94 in Development of fishes of the mid-Atlantic bight. An atlas of egg, larval, and juvenile stages, volume III. U.S. Fish and Wildlife Service, Washington, D.C. Harper, J. L., and H. E. Namminga. 1986. Fish population trends in Texoma Reservoir following establishment of striped bass. Pages 156–165 in G. E. Hall and M. J. Van Den Avyle, editors. Reservoir fisheries management: strategies for the 80’s. American Fisheries Society, Southern Division, Bethesda, Maryland. Harrell, R. M., editor. 1997. Striped bass and other Morone culture. Developments in Aquaculture and Fisheries Science 30. Harrell, R. M., J. H. Kerby, and R. V. Minton. 1990. Culture and propagation of striped bass and its hybrids. American Fisheries Society, Southern Division, Bethesda, Maryland. Harrell, R. M., and L. C. Woods. 1995. Comparative fatty acid composition of eggs from domesticated and wild striped bass (Morone saxatilis). Aquaculture 133:225–233. Hoar, W. S., and D. J. Randall, editors. 1988. Pages 15– 23 in Fish physiology, volume X. Academic Press, New York. Mansueti, R. 1958. Eggs, larvae, and young of the striped bass, Roccus saxatilis. University of Maryland, Chesapeake Biological Laboratory, Contribution 112, Solomons. McNair, H. M., and E. J. Bonelli. 1968. Basic gas chromatography. Consolidated Printers, Berkeley, California. Patrick, W. S., and M. L. Moser. 2001. Potential competition between hybrid striped bass (Morone saxatilis x M. americana) and striped bass (M. saxatilis) in the Cape Fear River estuary, North Carolina. Estuaries 24:425–429. Rainuzzo, J. R., K. I. Reitan, and Y. Olsen. 1997. The significance of lipids at early stages of marine fish: a review. Aquaculture 155:103–115. Regan, D. M., C. L. Wellborn, and R. G. Bowker. 1968. Striped bass development of essential requirements for production. Bureau of Sport Fisheries and Wildlife, Special Report, Atlanta. Rulifson, R. A., and R. W. Laney. 1999. Striped bass

stocking programs in the United States: ecological and resource management issues. Fisheries and Oceans Canada, Canadian Stock Assessment Secretariat Research Document 99/07, Ottawa. Rulifson, R. A. and C. S. Manooch, III. 1990. Recruitment of juvenile striped bass in the Roanoke River, North Carolina, as related to reservoir discharge. North American Journal of Fisheries Management 10:397–407. Rulifson, R. A., and K. A. Tull. 1999. Striped bass spawning in a tidal bore river: the Shubenacadie Estuary, Atlantic Canada. Transactions of the American Fisheries Society 128:613–624. Sargent, J. R. 1995. Origins and functions of egg lipids: nutritional implications. Pages 353–372 in N. Bromage and R. J. Roberts, editors. Brood stock management and egg and larval quality. Blackwell Scientific Publications, Oxford, UK. SAS Institute. 1999. The SAS software system for Windows version 8.2. SAS Institute, Cary, North Carolina. Seaborn, G. T., M. L. Jahncke, and T. I. J. Smith. 2000. Differentiation between cultured hybrid striped bass and wild striped bass and hybrid bass using fatty acid profiles. North American Journal of Fisheries Management 20:618–626. Setzler, E. M., W. R. Boynton, K. V. Wood, H. H. Zion, L. Lubbers, N. K. Mountford, P. Frere, L. Tucker, and J. A. Mihursky. 1980. Synopsis of biological data on striped bass, Morone saxatilis, (Walbaum). NOAA Technical Report NMFS Circular 433. Smith, H. M. 1896. A review of the history and results of the attempts to acclimatize fish and other water animals in the Pacific states. U.S. Fish Commission Bulletin 15(1895):379–472. Shulman, G. E., and R. M. Love. 1999. The biochemical ecology of marine fishes. Advances in marine biology, volume 36. Academic Press, San Diego, California. Talbot, G. P. 1966. Estuarine environmental requirements and limiting factors for striped bass. Pages 37–49 in R. F. Smith, A. H. Swartz, and W. H. Massmann, editors. A symposium on estuarine fisheries. American Fisheries Society, Special Publication 3, Bethesda, Maryland. USEPA (United States Environmental Protection Agency). 1987. Handbook of methods for acid deposition studies: laboratory analysis for surface water chemistry. USEPA, EPA 600/4–87/026, Washington, D.C. Worthington, R. E., and R. T. Lovell. 1973. Fatty acids of channel catfish (Ictalurus punctatus): variance components related to diet, replications within diets, and variability among fish. Journal of the Fisheries Board of Canada 30:1604–1608.