Toxicity Testing of Crude Oil and Related Compounds Using Early Life Stages of the Crimson-Spotted Rainbowfish (Melanotaenia fluviatilis)

Toxicity Testing of Crude Oil and Related Compounds Using Early Life Stages of the Crimson-Spotted Rainbowfish (Melanotaenia fluviatilis)

Ecotoxicology and Environmental Safety 52, 180}189 (2002) Environmental Research, Section B doi:10.1006/eesa.2002.2190 Toxicity Testing of Crude Oil ...

354KB Sizes 0 Downloads 16 Views

Ecotoxicology and Environmental Safety 52, 180}189 (2002) Environmental Research, Section B doi:10.1006/eesa.2002.2190

Toxicity Testing of Crude Oil and Related Compounds Using Early Life Stages of the Crimson-Spotted Rainbowfish (Melanotaenia fluviatilis) Carmel A. Pollino and Douglas A. Holdway Department of Biotechnology and Environmental Biology, RMIT-University, Bundoora Campus, GPO Box 71, Bundoora 3083, Australia Received January 18, 2001

The toxicity of petroleum hydrocarbons to marine aquatic organisms has been widely investigated; however, the e4ects on freshwater environments have largely been ignored. In the Australian freshwater environment, the potential impacts of petroleum hydrocarbons are virtually unknown. The toxicity of crude oil and related compounds were measured in the sensitive early life stages of the crimson-spotted rainbow5sh (Melanotaenia fluviatilis). Waterborne petroleum hydrocarbons crossed the chorion of embryonic rainbow5sh, reducing survival and hatchability. Acute exposures resulted in developmental abnormalities at and above 0.5 mg/L total petroleum hydrocarbons (TPH). Deformities included pericardial edema, disturbed axis formation, and abnormal jaw development. When assessing the acute toxicities of the water-accommodated fraction (WAF) of crude oil, dispersants, dispersant}oil mixtures, and naphthalene to larval rainbow5sh, the lowest to highest 96-h median lethal concentrations for day of hatch larvae were naphthalene (0.51 mg/L), dispersed crude oil WAF (DCWAF)-9527 (0.74 mg/L TPH), WAF (1.28 mg/L TPH), DCWAF-9500 (1.37 mg/L TPH), Corexit 9500 (14.5 mg/L TPH), and Corexit 9527 (20.1 mg/L). Using naphthalene as a reference toxicant, no di4erences were found between the sensitivities of larval rainbow5sh collected from adults exposed to petroleum hydrocarbons during embryonic development and those collected from unexposed adults.  2002 Elsevier Science (USA) Key Words: crude oil; dispersant; acute toxicity; early life stages; 5sh.

INTRODUCTION

The early life stages of "sh are particularly sensitive to xenobiotics (von Westernhagen, 1988). Acute toxicity experiments utilizing early life stages of "sh are often used to determine legally applicable measurements of pollutants and to measure their e!ects on aquatic biota (von

 To whom correspondence should be addressed at present address: Water Studies Centre, CRC for Freshwater Ecology, Monash University, Clayton Campus 3800, Vic., Australia. E-mail: ca}[email protected].

Westernhagen, 1988). Several studies have measured the acute toxicities of petroleum hydrocarbons to early and adult stages of "sh; however, few have measured multigenerational toxicity. Field studies conducted after the Exxon
180 0147-6513/02 $35.00  2002 Elsevier Science (USA) All rights reserved.

181

TOXICITY TESTING USING EARLY LIFE STAGES OF RAINBOWFISH

TABLE 1 LC50 Values for Crude Oil Exposure type Static Prudhoe Bay crude oil

Flow through Prudhoe Bay crude oil Static crude oil Continuous crude oil

Common name

Species

Lifecycle stage

Topsmelt

Atherinops a.nis

Larvae

96

16.34}40.20 mg/L Singer et al., 1998

Dolly Varden Chinook salmon Coho salmon Paci"c herring

Salvelinus malma Oncorhynchus tshawytscha Orcorhynchus kisutch Clupea pallasi

Juvenile Juvenile Juvenie Juvenile

96 96 96 96

1.23 mg/L 1.47 mg/L 1.45 mg/L 53.3 g/L

Moles et al., 1979 Moles et al., 1979 Moles et al., 1979 Carls et al., 1999

Lisa

Mugil curema

Adult

24

2400 mg/L

Templeton et al., 1975

Fathead minnow

Pimephales promelas

Adult

24

12.4 L/L

Hedtke et al., 1982

In marine environments, oil dispersants are often used to prevent spilled oil from reaching shorelines, thus not eliminating the problem but reducing the overall environmental impact. The two types of second/third-generation oil dispersants stockpiled in Australia and used in this study were Corexit 9527 and Corexit 9500 (Exxon, TX). Corexit 9527 is composed of 48% nonionic surfactants (ethoxylated sorbitan monooleate, ethoxylated sorbitan trioleate, and sorbitan monooleate), 35% anionic surfactant (sodium dioctyl sulfosuccinate), and 17% hydrocarbons solvent (ethylene glycol monobutyl ether) (Wolfe et al., 1998). Corexit 9500 is a hydrocarbon-based dispersant and can be used on light, heavy, and medium oils (Gilbert, 1996). It has the same surfactants as Corexit 9527, but provides enhanced penetration and emulsi"cation properties. Chemical dispersants alter the normal behavior of petroleum hydrocarbons by increasing their functional water solubility, resulting in increased bioavailability and altered interactions between dispersant, oil, and biological membranes (Wolfe et al., 2001). When comparing the toxicities of oil alone, dispersant alone, and oil and dispersant mixtures in the juvenile rain-

Duration (h)

LC 

Reference

bow trout (Oncorhynchus mykiss) (Lockhart et al., 1996), the oil and dispersant mixture was found to be more toxic as body water content was altered. While oil dispersants can be used in freshwater environments, little is known about the toxicities of Corexit 9527 and 9500 to freshwater organisms. When breaking down the toxicity of the components of crude oil, the aliphatic fraction is considered to be relatively nontoxic compared to the polyaromatic hydrocarbon components (PAHs), with small alkyl and aromatic hydrocarbons, principally the di- and tri aromatic compounds, believed to account for the majority of acute toxicity (Anderson, 1974). Naphthalene is a diaromatic hydrocarbon, a major component of crude oil and produced water (by-product of oil and gas production) (Gavin et al., 1996). Crude oils have been found to contain approximately 1.5% PAHs, of which 65% is made up of the naphthalenes (Truscott et al., 1992). Previous studies with naphthalene have found that the early life stages of "sh species are more sensitive to naphthalene exposure than adults (Table 2). This study used the crimson-spotted rainbow"sh (Melanotaenia -uviatilis), which are native to southeast Australia, and are found in the inland Murray}Darling system

TABLE 2 96-Hour LC50 Values for Naphthalene Common name

Species

Rainbow trout Largemouth bass Blue-gill Fathead minnow

Oncorhynchus mykiss Micropterus salmoides ¸epomis macrochirus Pimephales promelas

Sheepshead minnow Rainbow trout Fathead minow

Cyprinodon variegatus Oncorhynchus mykiss Pimephales promelas

Lifecycle stage Eggs and larvae Eggs and larvae Juvenile Adult Juvenile Adult Adult Adult

LC (mg/L)  0.12 0.68 15.2 6.1 1.99 2.4 1.6 7.9

Reference Rowe et al., 1983 Buccafusco et al., 1981 Holcombe et al., 1984 Anderson, 1974 DeGraeve et al., 1982 Holcombe et al., 1984

182

POLLINO AND HOLDWAY

TABLE 3 Mean (SE) pH, Temperature (3C), Dissolved Oxygen (mg/L), and Conductivity (S) (n ⴝ 9 for Each Measurement) for Embryonic and Larval Exposures Life stage

Exposure type

Embryo Larvae

WAF WAF Corexit 9527 Corexit 9500 DCWAF (9527) DCWAF (9500) Naphthalene

pH 7.2 6.9 6.9 7.2 7.1 7.2 7.1

(0.02) (0.06) (0.04) (0.01) (0.02) (0.01) (0.04)

(Allen, 1991). They are an ideal model test species for toxicity studies, as they can be reared in the laboratory. Rainbow"sh have been previously in toxicity testing (Holdway et al., 1994; Barry et al., 1995; Williams and Holdway, 2000), with newly hatched larvae being considered the most sensitive stage of the life cycle (Barry et al., 1995). The aims of this study were: (a) to measure the acute toxicity of embryonic rainbow"sh to the water-accommodated fraction of crude oil (WAF); (b) to compared acute toxicities of WAF, dispersants (Corexit 9527 and Corexit 9500), DCWAF (Corexit 9527 and Corexit 9500), and naphthalene to larval rainbow"sh; (c) to compare naphthalene 96-h LC values of larvae  collected from adult rainbow"sh exposed to one of untreated water, WAF, DCWAF (Corexit 9500), and naphthalene during their embryonic period. MATERIALS AND METHODS

Solution Preparation The WAF was prepared by mixing oil and water at a 1:9 ratio of oil to water for 24 h. After settling for 1 h, the aqueous fraction was collected and diluted. A volume of 250 mL of exposure solutions was extracted into 40 mL of dichloromethane (DCM) and analyzed for total petroleum hydrocarbons (TPH) using gas chromatography (GC). The dispersed crude oil water-accommodated fraction (DCWAF) was prepared by mixing a ratio of 1 : 9 oil to water for 24 h, removing crude oil from the top of the mixing chamber, and applying the dispersant Corexit 9527 or Corexit 9500 at a ratio of 1:50 dispersant to oil. This ratio was within the application guidelines and allows a good balance between hydrocarbons and dispersant (Gilbert, 1996). A 1-mL sample of this mixture was added to 1 L of WAF and mixed for a further 20 min. After settling, the aqueous fraction was collected and diluted. Volumes of 250 mL of exposure solutions were extracted into 40 mL of DCM and analyzed on GC for TPH. Dispersants were mixed directly into the test containers, and test concentrations were measured to 234 nm using

Dissolved oxygen 7.3 6.6 7.0 6.6 6.4 7.4 7.6

(0.1) (0.1) (0.2) (0.1) (0.2) (0.1) (0.1)

Conductivity 112 113 101 116 116 114 117

(3) (2) (2) (1) (2) (1) (1)

Temperature 24.4 24.3 24.3 24.0 24.9 24.9 24.8

(0.06) (0.02) (0.10) (0.04) (0.07) (0.07) (0.10)

a spectrophotometer (Cary 50, Varian) (Singer et al., 1994). Exposure concentrations were measured at the commencement of the experiment, before solution changes, and at the termination of the exposure. Naphthalene was dissolved in absolute ethanol (0.1 mL/L), meeting ASTM guidelines (ASTM, 1989). A constant amount of carrier was applied to carrier controls (CC) and exposure concentrations. Exposure concentrations were measured at the commencement of the experiment, before solution changes, and at the termination of the exposure. Water samples were measured at 220 nm using a spectrophotometer (Cary 50, Varian) (Stene and Lonning, 1985). Temperature, pH, dissolved oxygen, and conductivity were measured daily (Table 3). 96-Hour Embryonic Exposure Thirteen aquaria with three adult male and seven adult female rainbow"sh were supplied with arti"cial breeding substrates. After 1 h, eggs were collected by hand and treated with 4 mg/L malachite green for 5 min to prevent fungal growth (Holdway et al., 1994). Eggs were randomly distributed to glass Petri dishes and exposed to the WAF of Bass Strait crude oil for 96 h. Each Petri dish had 30 fertilized eggs. The exposure was conducted in triplicate. Replicates were commenced on consecutive days, as egg production was limited. Nominal WAF exposure concentrations were 0, 6.3, 12.5, 25, 50, and 100%, which corresponded to measured concentrations of 0 (0), 0.2 (0.03), 0.5 (0.03), 0.9 (0.06), 1.9 (0.09), and 3.8 (1.4) mg/L TPH. Petri dishes were randomly placed in an incubator maintained at 253C. Solutions in Petri dishes were completely replaced after each 24-h period. After exposure, eggs were transferred to #oating cups in aquaria supplied with toxicant-free water with a #ow su$cient to ensure to least 99% molecular replacement within 24 h (Sprague, 1969). Throughout the exposure, the developmental stages of the embryos were noted, along with any abnormalities.

TOXICITY TESTING USING EARLY LIFE STAGES OF RAINBOWFISH

183

TABLE 4 Nominal and Measured Concentrations of Exposure Solutions from Larval 96-hour LC50 Exposures Exposure type WAF Corexit 9527 Corexit 9500 DCWAF (9527) DCWAF (9500) Naphthalene

Nominal concentrations 0, 0, 0, 0, 0, 0,

5, 10, 20, 40, 80% 6.3, 12.5, 25, 50, 100 mg/L 1, 3.2, 10, 32, 100 mg/L 16, 3.1, 6.3, 12.5, 25% 0.5, 1.0, 2.0, 3.0, 6.0% CC, 0.32, 0.56, 1.0, 3.2, 5.6 mg/L

Observations continued through to hatching. Upon hatching, larval lengths were measured using a Wild binocular microscope, and any developmental abnormalities noted. Photographs were taken using a Wild binocular microscope and Leica photographic equipment. 96-Hour larval LC50 Exposures First-generation larval exposures. Arti"cial breeding substrates were placed in aquaria containing three male and seven female adult rainbow"sh. After 2 days, substrates were collected and maintained at 253C until hatching. On the day of hatching, larvae were collected from aquaria. Ten larvae were randomly assigned to exposure concentrations (Sprague, 1969). All exposures were conducted in triplicate. Rainbow"sh larvae were continually exposed to the controls or carrier controls (naphthalene only) or one of "ve concentrations of WAF, Corexit 9527, Corexit 9500, DCWAF (9527), DCWAF (9500), or naphthalene (Table 4) for 96 h. Larvae were counted at 24-h intervals for 96 h. Fifty percent of solutions were replaced at 24-h intervals. Second-generation larval exposures. All tests were conducted in triplicate. Eggs were collected from adult rainbow"sh that were exposed for 72 h during their embryonic period to one of "ve concentrations of WAF or DCWAF (9500), or one of seven concentrations of naphthalene. For each treatment group, 70 1-day-old larvae were collected. The nominal naphthalene exposure concentrations used in acute toxicity tests were 0, ethanol carrier control, 0.32, 0.56, 1.0, 3.2, and 5.6 mg/L, which corresponded to measured concentrations of 0, 0, 0.1 (0.05), 0.3 (0.03), 0.4 (0.05), 1.0 (0.05), and 2.2 (0.05) mg/L. Ten larvae were randomly assigned to each exposure concentration. Larvae were counted at 24-h intervals for 96 h. Statistical Analyses Ninety-six-hour LC values were calculated using the  Spearman}Karber method from the computer program Toxcalc (Tidepool Scienti"c). An analysis of variance (ANOVA) was used to determine statistical signi"cance of

Measured concentrations (mg/L) 0 0 0 0 0 0

(0), (0), (0), (0), (0), (0),

0.2 (0.03), 0.4 (0.07), 0.7 (0.06), 1.4 (0.6), 2.7 (0.9) TPH 5.9 (0.6), 11.4 (0.3), 21.0 (0.8), 48.3 (1.2), 92.0 (1.5) 0.9 (0.08), 2.1 (0.1), 9.8 (0.8), 38.5 (1.9), 87 (1.2) 0.3 (0.02), 0.7 (0.2), 1.3 (0.2), 2.5 (1.0), 5 (1.1) TPH 0.7, 1.3 (0.8), 2.5 (0.8), 5 (0.9), 10 (1.1) TPH 0 (0), CC, 0.1 (0.05), 0.3 (0.2), 0.5 (0.2), 0.9 (0.3), 1.9 (0.6)

individual biological responses. One-way and multi-way ANOVAs were used in analyses. If the ANOVA rejected a multisample null hypothesis of equal means (P(0.05), then Dunnett's test was used to determine signi"cance between samples (Zar, 1999). Tests for homogeneity were conducted using Bartlett's test or Levene's test. Data were transformed if results were not homogenous. Data transformations used were log to the base-10, natural log, or square root, and proportional data were transformed by arcsine square root as advised in Zar (1999). Statistics were performed using Statistica statistical software (Statsoft). RESULTS

Embryonic 96-Hour Exposure At the termination of the 96-h WAF exposure, the "rst developmental abnormalities and mortalities of the embryos were observed (Figs. 1A and 1B). The most common abnormality was an enlarged heart (pericardial edema) (Table 5). A!ected embryos developed at a slower rate than the control group. As development progressed, the incidence of deformities and mortalities increased (Figs. 2A and 2B). Abnormalities observed toward the end of the embryonic development included an enlarged pericardial sac, abnormal development of the jaw, and spinal deformities (Figs. 3A}3C). In extreme cases, the blood of the embryo was not pigmented. The majority of abnormalities were observed at 0.9 and 1.9 mg/L TPH, although there was a small incidence of abnormalities at the lower concentration (Table 5). Hatching commenced on Day 8. The predominant abnormalities of larvae were enlarged hearts, abnormal jaw development, and the vertebrae of larvae having slight #exures to bends (Figs. 4A}4C). Hatchability declined with increasing concentration, with no hatching at 3.8 mg/L due to 100% mortality in all replicates (Fig. 5). No changes in larval length were measured (Fig. 6). Larval 96-Hour LC50 Values First-generation larval exposures. The LC values at  24, 48, 72, and 96 h were calculated for all test compounds

184

POLLINO AND HOLDWAY

TABLE 5 Mean (SE) Incidence of Each Deformity (%) in Larval Rainbow5sh Exposed to WAF for 96 Hours as One-Hour-Old Embryos TPH (mg/L) 0 0.2 0.5 0.9 1.9 3.7

Pericardial edema

Vertebral abnormality

Jaw abnormality

0 0 16 (9) 18 (9)* 83 (16)* 0?

0 0 2 (2) 5 (4) 53 (15) 0?

0 0 3 (2) 5 (3) 17 (17)* 0?

Note. Values within treatment groups with an asterisk are signi"cantly di!erent from the control (P(0.05). ? 100% mortality of embryos.

abnormalities that a!ect survival were observed prior to and after hatching. Petroleum hydrocarbons are known to disrupt the development of embryonic "sh via multiple mechanisms, including altering cleavage membrane

FIG. 1. (A) Normal embryo at 96 h and (B) 96-h-old embryo with enlarged heart.

(Table 6). Day of hatch larvae were not highly sensitive to the dispersants Corexit 9527 and Corexit 9500 (Table 6). The 96-h LC values for WAF, DCWAF (9527), and  DCWAF (9500) were not signi"cantly di!erent (P'0.05) (Table 6). Compared to other toxicant exposures, larval rainbow"sh had the lowest 96-h LC for naphthalene.  Second-generation larval exposures. There were large con"dence limits for day of hatch naphthalene LC values  for larvae from parents exposed as embryos (Table 7). Comparisons within or between treatment types were not statistically signi"cant (P'0.05). DISCUSSION

Embryonic Exposure Although WAF was not acutely toxic to embryonic rainbow"sh during the 96-h exposure period, developmental

FIG. 2. (A) Normal embryo at Day 6 and (B) 6-day embryo with enlarged heart and slow larval development as shown by the poor development of lenses.

TOXICITY TESTING USING EARLY LIFE STAGES OF RAINBOWFISH

185

FIG. 4. (A) Normal 1-day-old larvae, (B) larve with enlarged heart, and (C) poorly developed larvae with enlarged heart and abnormal jaw.

lenses, a bent notochord, and inhibited pigmentation (von Westernhagen, 1988). In this study, hatchability and the incidence of larval deformities were decreased at 0.5 mg/L TPH and above.

FIG. 3. (A) Normal embryo at Day 8, (B) 8-day embryo with enlarged heart, and (C) poorly developed 8-day embryo with enlarged heart and abnormal development of the lenses.

formation and blocking the phosphorylation of ADP (von Westernhagen, 1988). Petroleum hydrocarbons are also depressors of embryo activity and typical responses of exposed larva are poorly di!erentiated head, protruding eye

FIG. 5. Percentage hatch of larvae exposed for 96 h during the early embryonic period. Values with a letter in common are not signi"cantly di!erent (P'0.05).

186

POLLINO AND HOLDWAY

TABLE 7 96-Hour LC50 (95% Con5dence Limits) Values (g/L) of Day of Hatch Larvae Exposed to Naphthalene Parent exposure type Naphthalene

TABLE 6 24-, 48-, 72-, and 96-Hour LC50 Values (95% Con5dence Limits) for Day of Hatch LC50 (95% Con5dence Limits) Values for Day of Hatch Crimson-Spotted Rainbow5sh Exposed to Corexit 9527, Corexit 9500, WAF of Crude Oil, DCWAF (Corexit 9527), DCWAF (Corexit 9500), and Naphthalene (n ⴝ 4) for each LC50) Exposure

Time (h)

LC50 (mg/L)

WAF

24 48 72 96

4.48 3.38 2.10 1.28

(4.0, (2.8, (1.8, (1.0,

Corexit 9527

24 48 72 96

64.3 61.2 47.3 20.1

Corexit 9500

72 96

34.7 (30, 46) 14.5 (12, 18)

DCWAF (9527)

48 72 96

2.92 (2.0, 4.2) 1.25 (1.9, 1.5) 0.74 (0.5, 1.0)

DCWAF (9500)

24 48 72 96

2.62 1.94 1.67 1.37

Naphthalene

72 96

1.21 (0.8, 2.0) 0.51 (0.4, 0.7)

(36, (35, (34, (16,

5.2) 3.0) 2.6) 1.6)

116) 106) 66) 25)

(2.2, (1.7, (1.5, (1.2,

Note. WAF and DCWAF values given as mg/L TPH.

3.1) 2.2) 1.9) 1.5)

96-h LC 

Control Carrier control 0.03 0.05 0.09 0.28 0.50

1174 (824, 2299) 553 (360, 848) 372 (139, 703) 837 (245, 3600) 438 (186, 837) 686 (246, 2306) 961 (403, 1820)

WAF

Control 0.26 0.31? 0.44 1.03

213 (117, 311) 640 (309, 1306) * 313 (260, 376) 470 (262, 712)

DCWAF (9500)

Control 0.26 0.53 0.816 3.014?

705 (468, 1092) 520 (273, 816) 1171 (767, 7826) 315 (149, 510) *

FIG. 6. Lengths of larvae exposed to WAF for 96 h during the early embryonic period. Values with a letter in common are not signi"cantly di!erent (P'0.05).

Exposure of larval Paci"c herring (Clupea pallasi) to watersoluble fraction (WSF) of crude oil also a!ected hatching success, frequency of gross abnormalities, and larval lengths (Smith and Cameron, 1979), Baltic herring (Clupea harengus

Parental exposure concentration (mg/L)

Note. Larvae were collected from parental "sh exposed as embryos to naphthalene, WAF, and DCWAF (Corexit 9500) (n"3 for each LC ).  ? Unable to collect su$cient larvae to conduct LC . 

membras) embryos exposed to 50 g/L WSF had a reduction in hatchability (Linden, 1978), and embryonic walleye Pollock (¹heragra chalcogramma) exposed to WSF had reduced larval lengths (Carls and Rice, 1990). Larval lengths of rainbow"sh were not signi"cantly reduced, despite a negative concentration-dependent decline in length with TPH concentration being observed. The concentrationdependent reduction in larval length may be the result of stimulation of embryonic metabolism at the expense of growth, as suggested by Smith and Cameron (1979). Exposure of rainbow trout embryos (Oncorhynchus mykiss) to environmentally relevant concentrations of benzo[a]pyrene for only a short period resulted in abnormalities after emergence (Ostrander et al., 1990). The lipophilic nature of benzo[a]pyrene and the expansive yolk reserves resulted in the uptake of the chemical which was slowly lost from the egg through development. When exposing embryonic or larval stages to oil, the yolk sac is a likely storage point for hydrocarbons entering tissues during an exposure (Vandermeulen, 1984). The slow release of hydrocarbons during development could result in a chronic exposure to toxicants that prolong the manifestation of developmental abnormalities (Vandermeulen, 1984). In this study, the incidence of abnormalities in rainbow"sh larvae continued to increase in developing rainbow"sh after completion of the 96-h exposure, suggesting petroleum hydrocarbons may have accumulated in the yolk sac, although petroleum

TOXICITY TESTING USING EARLY LIFE STAGES OF RAINBOWFISH

hydrocarbons crossing the chorion and irreversible damage during the exposure period could also have contributed to the continuing manifestation of abnormalities beyond the exposure period. Ascites are histopathological lesions often observed in the early development stages of "sh after exposure to xenobiotics. These are characterized as #uid "lling the peritoneal cavity, causing pericardial edema or hydropericardium (Marty et al., 1997). Edema has previously been shown to be responsible for larval mortality (Carls et al., 1999). Pericardial edema was a common abnormality observed in rainbow"sh after exposure to WAF, and made up the majority of abnormalities at 1.9 mg/L TPH. The "rst incidences of pericardial edema were found after the 96-h exposure, with increasing incidence throughout the developmental period. Edema can also lead to delayed development (Marty et al., 1997). Edema and slow development of embryos appeared to coincide in this study, with an increase in the incidence of deformities and a decrease in larval length being observed with increasing concentration. Abnormal jaw development is common in oil-exposed larvae (von Westernhagen, 1988), with craniofacial and mandibular morphological malformations being observed in larval Paci"c herring after the Exxon
187

curema was greatest in the "rst 2 h of the exposure, with toxicity decreasing as volatiles evaporated (Templeton et al., 1975). Exposure solutions in this study were partially replaced throughout the toxicity tests, which may explain why rainbow"sh toxicities were not restricted to early in the exposure period, but increased over the 96 h. Exposures were not continued beyond 96 h as larval rainbow"sh switch from endogenous to exogenous feeding, and mortalities are high. Rainbow"sh sensitivity to WAF was comparable to those of other studies; however, di!erent toxicities of crude oil to "sh are expected due to the di!erent chemical compositions of oil types and di!erent experimental procedures. The larval rainbow"sh LC values for crude oil and the  two types of dispersed crude oil were not signi"cantly di!erent. A study by Wolfe et al. (2001) found that during the "rst 2 h of exposure to dispersed crude oil (Corexit 9527), naphthalene uptake in larval topsmelt (Atherinops a.nis) was enhanced, compared to crude oil alone. In the subsequent 10 h, naphthalene uptake was the same for exposure solutions and thus over extended periods hydrocarbon uptake by larvae may not be altered by the addition of the dispersant. This may account for the lack of di!erences in the toxicities of test solutions observed in this study. Another study exposing larval topsmelt to both crude oil and dispersed crude oil (Corexit 9527) found that although dispersed crude oil (96-h LC : 28.60}74.73 mg/L) was more toxic  than crude oil alone (96-h LC : 16.34}40.20 mg/L), the  data were highly variable (Singer et al., 1998). The di!erences in toxicity were believed to be due to the additional dissolved materials in exposure solutions, the physical e!ects associated with contact with the droplets, and the enhanced uptake resulting from animal-to-oil contact (Singer et al., 1998). Dispersants alone were not highly toxic to rainbow"sh and were thus unlikely to cause additional rainbow"sh mortality if used in an oil spill situation. Corexit 9500 alone was more toxic than Corexit 9527; however, this was reversed for dispersant}oil mixtures. The change in order of toxicities may be related to the di!erent degradation rates and/or degradation products of the dispersants. In a study using juvenile rainbow trout, "sh were more sensitive to dispersant}oil mixtures than oil or dispersant alone as body water content was altered, and mortality was more rapid (Lockhart et al., 1996). However, in Adams et al. (1999) dispersed crude oil was less toxic than either oil or dispersant alone, indicating that toxicity data vary for di!erent dispersants and di!erent oil types. Little information is available for the toxicities of dispersant solutions alone. These data suggest that Corexit 9500 and Corexit 9527 alone are not highly toxic to early life stages of rainbow"sh. In this study, rainbow"sh had the highest sensitivity to naphthalene. The majority of larval mortality occurred

188

POLLINO AND HOLDWAY

during the "nal 48 h of the exposure. The embryo}larval stages of the rainbow trout and large mouth bass (Micropterus salmoides) exposed to naphthalene from fertilization until 4 days after hatching had LC values of 0.11 mg/L for  rainbow trout and 0.51 mg/L for largemouth bass (Micropterus salmoides) (Black et al., 1983) which is comparable to the rainbow"sh LC value calculated in this study. The  majority of LC values for other species are higher than the  LC value of this study. Fathead minnow (Pimephales  promelas) were also less sensitive to naphthalene exposure, as concentrations over 0.85 mg/L naphthalene were needed to reduce hatchability and depress the length and weight of fry at 30 days (DeGraeve et al., 1982). Exposures to 4.38 and 8.51 mg/L of naphthalene reduced larval fathead minnow survival, with no fry surviving at 30 days (DeGraeve et al., 1982). Multigenerational e!ects after have been observed in natural populations of the pink salmon (Cronin and Bickham, 1998), paci"c herring (Kocan et al., 1996), and pink salmon (Adamas et al., 1999) following the Exxon
Following major spills of petroleum hydrocarbons into freshwater environments, concentrations of 4 mg/L TPH (Guiney et al., 1987) and 5 mg/L TPH (Caldwell, 1997) have been measured; thus exposure concentrations for WAF and DCWAF were environmentally relevant. The measured concentration of naphthalene in produced water has been reported to be 133 g/L, which is lower than the LC of  rainbow"sh, and therefore unlikely to cause acute toxicity but may cause sublethal damage to early life cycle stages. Future studies are required to investigate the e!ects of sublethal concentrations of crude oil and associated compounds on di!erent life stages of the rainbow"sh.

ACKNOWLEDGMENTS The authors acknowledge Exxon Australia for supplying Bass Strait crude oil and the dispersants Corexit 9500 and Corexit 9527. This research was supported by an Australian Postgraduate Award to C. Pollino and ARC Grant A00000256 to D.A. Holdway.

REFERENCES Adams, G., Klerks, P., Belanger, S., and Dantin, D. (1999). The e!ect of the oil dispersant Omni-clean on the toxicity of fuel oil No. 2 in two bioassays with the sheepshead minnow Cyprinodon variegates. Chemosphere 39, 2141}2157. Allen, G. (1991). Field Guide to Freshwater Fishes of New Guinea. Christensen Research Institute, Publication No. 9. Anderson, J. (1974). Responses to sublethal levels of petroleum hydrocarbons: Are they sensitive indicators and do they correlate to tissue contamination? In Methods for Chemical Analysis of =ater and =astes, ;.S. Environmental Protection Agency. Methods Development and Quality Assurance Research Laboratory, Cincinnati, pp. 95}114. ASTM. (1989). E 729-99a: Standard guide for conducting acute toxicity tests with "shes, macroinvertebrates, and amphibians. In American Society for ¹esting and Materials, pp. 403}422. ASTM, Philadelphia. Barry, M., Logan, D., van Dam, R., Ahokas, J., and Holdway, D. (1995). E!ect of age and weight-speci"c respiration rate on toxicity of esfenvelerate pulse-exposure to the Australian crimson-spotted rainbow "sh (Melanotaenia -uviatilis) Aquat. ¹oxicol. 32, 115}126. Black, J., Birge, W., Westerman, A., and Francis, P. (1983). Comparative aquatic toxicology of aromatic hydrocarbons. Fundam. Appl. ¹oxicol. 3, 353}358. Buccafusco, R., Ells, S., and LeBlanc, G. (1981). Acute toxicity of priority pollutants to bluegill (¸epomis macrochirus). Bull. Environ. Contam. ¹oxicol. 26, 446}452. Carls, M., Holland, L., Larsen, M., Lum, J., Mortensen, D., Wang, S., and Wertheimer, A. (1996). Growth, feeding, and survival of pink salmon fry exposed to food contaminated with crude oil. Am. Fish. Soc. Sympos. 18, 608}618. Carls, M., and Rice, S. (1990). Abnormal development and growth reductions of Pollock ¹heragra chalcogramma embryos exposed to water-soluble fractions of oil. Fish. Bull. 88, 29}37. Carls, M., Rice, S., and Hose, J. (1999). Sensitivity of "sh embryos to weathered crude oil: I. Low-level exposure during incubation causes malformations, genetic damage, and mortality in larval Paci"c herring (Clupea pallasi). Environ. ¹oxicol. Chem. 18, 481}493. Cronin, M. A., and Bicham, J. W. (1998). A population genetic analysis for a crude oil spill to induce heritable mutations and impact natural populations. Ecotoxicol. 7, 259}278. DeGraeve, G., Elsder, R., Woods, D., and Bergman, H. (1982). E!ects of naphthalene and benzene on fathead minnows and rainbow trout. Arch. Environ. Contam. ¹oxicol. 11, 487}490. Gavin, M., Brooke, D., Howe, P., and Dodson, S. (1996). Environmental hazard assessment: Naphthalene. In ¹SD/27. Chemicals and Biotechnology Division, Department of Environment, Garston, Watford. Gilbert, T. (1996). Facts on the composition, toxicity and use of oil spill dispersants. Sixth Scienti,c Support Coordinators =orkshop, 2}6 December 1996, ¸aunceston, ¹asmania. Hedtke, S., and Puglisi, F. (1982). Short-term toxicity of "ve oils to four freshwater species. Arch. Environ. Contam. ¹oxicol. 11, 425}430.

TOXICITY TESTING USING EARLY LIFE STAGES OF RAINBOWFISH Heintz, R., Short, J., and Rice, S. (1999). Sensitivity of "sh embryos to weather crude oil. II. Increased mortality of pink salmon (Oncorhynchus gorbuscha) embryos incubating downstream from weathered Exxon
189

Singer, M., George, S., Jacobson, S., Lee, I., Jacobson, S., Weetman, L., Blondina, G., Tjeerdema, R., Aurand, D., and Sowby, M. (1998). E!ects of dispersant treatment on the acute toxicity of petroleum hydrocarbons. Arch. Environ. Contam. ¹oxicol. 34, 177}187. Singer, M., George, S., Jacobson, S., Lee, I., Tjeerdema, R., and Sowby, M. (1994). Comparative e!ects of oil dispersants to the early life stages of topsmelt (Atherinops a.nis) and kelp (Macrocystis pyrifera). Environ. ¹oxicol. Chem. 13, 649}655. Smith, R. L., and Cameron, J. A. (1979). E!ect of water soluble fraction of Prudhoe Bay crude oil on embryonic development of Paci"c Herring. ¹rans. Am. Fish. Soc. 108, 70}75. Sprague, J. B. (1969). Measurement of pollutant toxicity to "sh. I. Bioassay methods for acute toxicity. =ater Res. 3, 793}821. Stene, A., and Lonning, S. (1985). E!ects of short-term exposures to naphthalene, methyl- and hydroxynaphthalenes on 2 di!erent embryonic stages of cod (Gadhus morhua L.). Sarsia 7, 279}285. Templeton, W., Sutton, E., Bean, R., Arnett, R., Blaylack, J., and Wildung, R. (1975). Oil pollution studies on lake Maracaibo, Venezeula. In 1975 Conference on Prevention and Control of Oil Pollution, Vol. 99, pp. 489}496. San Francisco. Truscott, B., Idler, D., and Fletcher, G. (1992). Alteration of reproductive steroids of male winter #ounder (Pleuronectes americanus) chronically exposed to low levels of crude oil in sediments. Can. J. Fish. Aquat. Sci. 49, 2190}2195. Vandermeulen, J. (1984). Toxicity and sublethal e!ects of petroleum hydrocarbons in freshwater biota. In Oil in Freshwater: Chemistry, Biology, Countermeasure ¹echnology (J. Vandermeulen and S. Hrudley, Eds.), pp. 267}303. Pergamon Press, New York. von Westernhagen, H. (1988). Sublethal e!ects of pollutants on "sh eggs and larvae. In Fish Physiology (W. Hoar and D. Randall, Eds.), pp. 253}345, Academic Press, San Diego. Williams, N. D., and Holdway, D. A. (2000). The e!ects of pulse-exposed cadmium and zinc on embryo hatchability, larval development, and survival of Australian crimson spotted rainbow "sh (Melanotaenia -uviatilis) Environ. ¹oxicol. 15, 165}173. Wolfe, M. F., Schlosser, J., Schwartz, G., Singaram, S., Mielbrecht, E., Tjeerdema, R., and Sowby, M. (1998). In#uence of dispersants on the bioavailability and trophic transfer of petroleum hydrocarbons to primary levels of a marine food chain. Aquat. ¹oxicol. 42, 211}227. Wolfe, M., Schwartz, G., Singaram, S., Mielbrecht, E., Tjeerdema, R., and Sowby, M. (2001). In#uence of dispersants on the bioavailability and trophic transfer of petroleum hydrocarbons to larval topsmelt (Atherinops a.nis). Aquat. ¹oxicol. 52, 49}60. Zar, J. H. (1999). Biostatistical Analysis. Prentice}Hall, New York.