.

4quaculture, 91 (1990)265-280 265

Elsevier Science Publishers B.V.. Amsterdam

Acute and chronic toxicity of ammonia to

juvenile Metapenaeus macleayi and
Penaeus wlonodon and the influence of low
dissolved-oxygen levels

Geoff L. Allan, Greg B. Maguire’ and Stephen J. Hopkins2

N.S. I+: Agriculture and Fisheries, Brackish Water Fish Culture Research Station,
Salamander Bay. N.S. I+: 2301. Australia
(Accepted 30 April 1990 )

ABSTRACT

Allan. G.L.. Maguire, G.B. and Hopkins, S.J., 1990. Acute and chronic toxicity of ammonia to juve-
nile Metapenaeus macleayi and Penaeus monodon and the influence of low dissolved-oxygen levels.
:lquaculture, 9 1: 265-280.

Acute toxicity of ammonia was estimated as 96-h LCso values. For juvenile school prawns, Meta-
penaeus macleayi, and leader prawns, Penaeus monodon. these were 1.39 and 1.69 mg un-ionised
ammonia, NH3-N/l (26.3 and 37.4 mg total ammonia-N/l). respectively. Reduced dissolved-oxygen
(DO) levels significantly (PcO.05) increased the acute toxicity of ammonia to P. monodon. Ninety
percent of prawns held for 96 h at a DO level of 2.3 mg/l and an ammonia concentration of 1.60 mg
NH,-N/I (33.5 mg total ammonia-N/l) died, whereas only 33.3% died at a DO level of 5.7 mg/l and
a similar ammonia concentration of 1.63 mg NHX-N/I (33.9 mg total ammonia-N/l). The “maxi-
mum acceptable” level of ammonia was defined as that which reduced growth by 5% over 3 weeks.
For M. macleayi and P. monodon these levels were 0.35 and 0.21 mg NH3-N/I (7.7 and 4.1 mg total
ammonia-N/l), respectively.

INTRODUCTION

Ammonia is very toxic to aquatic animals and can cause impairment of

cerebral energy metabolism and damage to gill, liver, kidney, spleen and thy-
roid tissue in fish, crustaceans and molluscs (Smart, 1978; Colt and Arm-
strong, 198 1). It can limit production in intensive fish and crustacean aqua-
culture (Delistraty et al., 1977; Colt and Armstrong, 198 1) and has killed fish
in the wild (Tarazona et al., 1987).

‘Present address: Key Centre for Teaching and Research in Aquaculture. Tasmanian State In-
stitute of Technology, Box 1214, Launceston, Tasmania 7248, Australia.
“Present address: Hunter District Water Board, Chemistry Section, Ravenshaw St., Newcastle,
N.S.W. 2300. Australia.

0044-8486/90/$03.50 0 1990 - Elsevier Science Publishers B.V.

266 G.L. ALLAN ET AL.

In solution total ammonia comprises un-ionised ammonia ( NH3) and ion-

ised ammonia (NH,+ ) in equilibrium. The proportions of each depend mainly
on pH, but also on temperature, salinity and pressure (Whitfield, 1974). NH:,
is by far the more toxic (Smart, 1978); however, NH: may also become
toxic, especially at low pH levels when the proportion of ammonia as NH: is
very high (Shaw, 1960; Armstrong et al., 1978).
Ammonia is the principle nitrogenous product excreted by crustaceans
( Claybrook, 1983 ) and may also accumulate in culture systems following mi-
crobial decomposition of organic material (Stanier et al., 1976) and with some
fertilisation practices (Boyd, 1982). As low dissolved-oxygen (DO) levels
can occur following organic decomposition, an interaction between ammonia
and DO could be important for aquaculturists.
Low dissolved-oxygen concentrations increase the toxicity of ammonia to
fish (Alabaster et al., 1979; Thurston et al., 198 1) and Lloyd ( 196 1) attrib-
uted this to an increase in the uptake of ammonia as gill ventilation rates
increased to prevent hypoxia. No reports describing the effects of low dis-
solved-oxygen levels on ammonia toxicity for crustaceans were found.
Previous studies on the toxicity of ammonia to crustacean larvae have in-
volved lobsters, Homarus americanus (Delistraty et al., 1977), freshwater
prawns, Macrobrachium rosenbergii (Armstrong et al., 1978) and penaeids,
Penaeus indicus (Jayasankar and Muthu, 1983) and P. monodon (Chin and
Chen, 1987). One study used juvenile penaeids ( < 2 g) but the results for
several species were pooled (Wickins, 1976). All used static bioassay meth-
ods to estimate critical levels of ammonia. This approach is limited, as ani-
mals are stressed when test solutions are replaced, and metabolic wastes can
increase the concentrations of ammonia between water exchanges, especially
when larger animals are used and in chronic bioassays when animals are fed.
Most studies of ammonia toxicity involve a single species. Comparisons
between species are made using results from different bioassay systems; how-
ever, comparisons are more valid if only one bioassay system is used. In the
present study continuous-flow bioassays were used to determine the acute and
chronic toxicity of ammonia to juvenile school prawns, Metupenaeus ma-
cleayi, and leader prawns, P. monodon, both commonly farmed in Australia
(Maguire and Allan, 1985; Maguire et al., 1988). The effects of low dissolved
oxygen on the acute toxicity of ammonia to leader prawns were also investi-
gated. Results from this study may assist with the management of prawn-cul-
ture systems especially intensive nursery and grow-out units which can expe-
rience high ammonia levels (Chen et al., 1988).

MATERIALS AND METHODS

Leader prawns, P. monodon, came from prawn-farming ponds (Goodwood

Island, N.S.W.) and school prawns, Metapenaeus macleuyi, were otter trawled
TOXICITY OF AMMONIA TO PRAWNS 267

from Port Stephens, N.S.W. They were acclimatised in the experimental

aquaria for at least a week before experiments commenced. During both ac-
climation and chronic bioassays, prawns were fed freshly shucked and chopped
bivalve flesh (pipi, Plebidonax de/&ides).

Bioassay system
Continuous-flow bioassays were conducted in 18 acrylic aquaria (70 1) with
ten prawns in each. All experiments had six treatments with three randomly
assigned replicate aquaria per treatment. Except for the low dissolved-oxygen
experiment (Experiment 3 ) treatments included one control, with no added
ammonia, and five with different ammonia concentrations. Concentrated
ammonia solutions were prepared in 20-l reservoirs and pumped at 2.5 ml/
min to 500 ml mixing flasks, diluted by flowing seawater and then supplied
to the aquaria. Seawater was filtered through a commercial sand filter and 5-
,um and 1-pm cartridge filters before entering the mixing flasks at a controlled
flow rate of 245 ml/min. The combined flow (ammonia solution plus sea-
water) of 247.5 ml/min replaced approximately 90% of each 70-l aquarium
in 10 h as recommended by Sprague ( 1969). Except for Experiment 3, sea-
water was vigorously aerated before reaching the mixing flasks. Ammonium
chloride (NH&l) was used to prepare ammonia solutions, and pH was ad-
justed in the 20-l reservoirs to that of incoming seawater by the addition of
10 M NaOH.
No sediment was provided for the prawns because it reduced the concen-
tration of ammonia, presumably due to nitrification. Control of pH was also
harder when sediment was provided. The absence of sediment did not in-
crease cannibalism among school prawns starved over 96 h or significantly
affect growth (P> 0.05 ) in 3-4 weeks when school or leader prawns were fed
ad libitum with pipi flesh (Allan and Maguire, unpublished data, 1989). Sea-
water was maintained at constant temperatures of 25 ? 1‘C and 27 ? 2 ‘C for
bioassays with school and leader prawns, respectively. These temperatures
were within the optimum range for growth of each species (Maguire and Al-
lan, unpublished data , 1989). A 12: 12 hour photoperiod of light: dark was
maintained. Except for Experiment 3, dissolved-oxygen levels were main-
tained above 5.5 mg/l in all experiments either by vigorously aerating incom-
ing water (Experiments 1 and 2) or by using two airstone diffusers in each
aquarium (Experiments 4 and 5),
The bioassay seawater was essentially oceanic with slightly elevated levels
of suspended solids as described by Scribner ( 1986). Total nitrite-N plus ni-
trate-N levels in acute bioassays ( l-3 ) remained below 30 pg/l, but in chronic
bioassays (4-5 ), levels reached 111 pg/l, still well below those which reduced
growth of juvenile penaeids (Wickins, 1976).
268 G.L. ALLAN ET AL.

Water quality analyses

The ammonia concentration, pH level and temperature in each aquarium
were measured daily. The indophenol blue colourmetric method (Dal Pont
et al., 1973) was used to measure ammonia. Un-ionised ammonia concentra-
tions, measured as ammonia nitrogen (NH,-N) were calculated daily from
measurements of total ammonia nitrogen (total ammonia-N), pH, tempera-
ture and salinity (Bower and Bidwell, 1978). Orion and Metrohm pH/mV
meters, with Ross combination and Metrohm reference and glass electrodes
respectively, were used to measure pH. The electrodes were calibrated with
phosphate and borate buffers (American Public Health Association, 197 1)
and both systems agreed ( -+ 0.01 pH units). Nitrite and nitrate were mea-
sured occasionally using colourmetric methods (Major et al., 1972 ). Salinity
was measured regularly using a Yeo-Kal temperature/salinity conductivity
meter (Yeo-Kal Electronics, Brookvale, Sydney, N.S.W. 2 100, Australia) cal-
ibrated with substandard seawater and a standard thermometer. Dissolved
oxygen was measured with a Yeo-Kal dissolved oxygen/temperature meter
calibrated before each set of measurements with “air-saturated” seawater and
periodically using Winkler’s titration (A.P.H.A., 197 1).

Experiments I and 2 - Acute bioassays

To assess the acute toxicity of ammonia to school and leader prawns, six
experimental levels ranging from 0.003-3.54 mg NH,-N/l (0.06-66.4 mg to-
tal ammonia-N/l) and 0.002-2.27 mg NH,-N/l (0.04-5 1.O mg total ammo-
nia-N/l) respectively, were established. Just before the start of each acute
bioassay, ammonia and pH levels in each aquarium were quickly adjusted
with ammonium chloride and sodium hydroxide. The initial mean weights
for school and leader prawns were 2.0 g (range 0.9-4.8) and 2.2 g (range 0.8-
4.5 ), respectively. Prawns were not sexed for the acute bioassays, no aeration
was provided and prawns were not fed. Absence of response to touching with
a glass rod was the criterion of death and all dead prawns were removed. Fol-
lowing A.P.H.A. ( 197 1) and Sprague ( 1969), the acute bioassays were run
for 96 h. Longer periods were not considered appropriate due to stress and
cannibalism. Mortality checks were made after 0.5, 1, 2 and 4 h and at four
hourly intervals thereafter.

Experiment 3 - Effect of low dissolved oxygen on acute ammonia toxicity

In this experiment only unsexed leader prawn were used and, except for
oxygen control, similar methods to those described for Experiments 1 and 2
were used. Average initial prawn weight was 4.2 g (range 2.0-8.3). The dis-
solved-oxygen concentration was reduced for five of the six treatments to ap-
proximately 2.2 mg/l using a vacuum, gas-stripping apparatus similar to that
described by Mount ( 196 1, 1964). One of these five treatments was a control
(no added ammonia) and in the other four ammonia was added to give 0.53,
TOXICITY OF AMMONIA TO PRAWNS 269

0.95, 1.30 and 1.63 mg NH,-N/l ( 12.6, 19.9, 27.0 and 33.9 mg total ammo-
nia-N/l). For the sixth treatment vigorously aerated seawater was used and
ammonia was added to give 1.60 mg NH,-N/I.

Experiments 4 and 5 - Chronic bioassays

To assess the effects of ammonia on weight gain, survival and food conver-
sion efficiency for school and leader prawns, six experimental levels ranging
from 0.01-l .20 mg NH,-N/l (0.2-24.2 mg total ammonia-N/l) and 0.02-
1.60 mg NH,-N/l (0.3-32.9 mg total ammonia-N/l) respectively, were es-
tablished. For Experiment 4 school prawns averaged 2.5 g (range 0.9-5.6)
and for Experiment 5 leader prawns averaged 3.7 g (range 1.6-6.9). Prawns
were individually tagged before stocking by injecting a marker of petroleum
jelly impregnated with a fluorescent dye, “saturn yellow”, into the muscula-
ture of one of the abdominal segments on either side (Klima, 1965 ) . Prawns
were fed ad libitum, and gentle aeration was provided throughout. As juvenile
male and female school prawns grow at different rates (Maguire and Allan,
1985 ), equal numbers of males and females were stocked into each aquarium
for Experiment 4, but as leader prawns grow at similar rates when < 13 g
(Liao, 1977), prawn sex was not recorded for Experiment 5. At the start of
each chronic bioassay, ammonia concentrations were allowed to rise gradu-
ally, reaching required concentrations within 48 h.
After 3 weeks exposure, the average individual prawn weight gain was de-
termined for each aquarium. Weight gain values were based on individual
measurements of initial and final weights for prawns that survived or that
died during the last 3 days of an experiment. The minimum numbers of
prawns, in any aquarium, used for average individual weight gain determi-
nations were n 2 6 for school prawns and n 3 7 for leader prawns.

Statistical analyses
For acute bioassays (Experiments 1 and 2) mortality data represent deaths
over 96 h unless stated. The dose response at each concentration was deter-
mined with probit analysis (Busvine, 1957 ), using pooled data for replicate
aquaria. Acute toxicity was expressed as 96-h L&, that concentration which
killed 50% of the prawns in 96 h. To enable comparisons with other studies,
48 h LCsO values were also calculated.
For all other experiments (3-5), treatment effects were identified using
analysis of variance (ANOVA), and multiple comparisons among means were
made using Tukey’s honestly significant difference technique (Sokal and
Rohlf, 198 1) . Homogeneity of variance was confirmed using Cochran’s test
(Winer, 197 1). Throughout the paper mean 2 s.e. values are given, except
where 95% confidence limits are indicated.
For the chronic bioassay with school prawns (Experiment 4) two-factor
 G.L. ALLAN ET AL.

013 016 019 112

NH, (m9-N/Q

Fig. 1. Effects of ammonia on growth of school prawns (Metupenaeus macleayi) over 2 I days
(Experiment 4).

ANOVA was used to investigate the effects of ammonia concentration and

sex on growth. For all other analyses, including the effect of ammonia concen-
tration on leader prawn growth (Experiment 5 ) , single-factor ANOVA’s were
used. Survival and food conversion efficiency for each aquarium were also
calculated. The food conversion ratio (FCR) was estimated by dividing the
total amount of food eaten by the prawns in each aquarium by the total prawn
biomass increase in each aquarium (g wet weight). Weight of bivalve fed was
adjusted to 92% dry weight for comparisons with values recorded in trials
with pelleted artificial diets (Maguire et al., 1988). In aquaria where mortal-
ity occurred, the initial weights of the prawns that died were not considered
in FCR calculations. When survival was less than 80% within an aquarium,
the data for that treatment were not included in statistical analysis of FCR.
To meet the assumptions of normality and homogeneity of variance, FCR
data were transformed (log X) prior to analysis.
The EC5 value, that concentration at which prawn growth was reduced by
5%, was considered the “maximum acceptable” level of ammonia. This was
TOXICITY OF &MMONIA TO PRAWNS 271

Y = -0.823X + 3.881

I
I
I
I Y=-3.623X + 4.324
I
I
I
I
I
I
I
I
I
I
I
I
I
/
,
I
I
I
I
/ EC,= 0.21mg NH,-N/I
I

t 1 1
0.3 0.6 0.9 1:2
NH, (mg-N/0

Fig. 2. Effects of ammonia on growth of leader prawns (Penaeus r~~~odorz) over 2 I days (Ex-
periment 5 ).

estimated from two intersecting linear regressions (Sedgwick, 1979; Maguire

and Hume, 1982). The first regression line was based on the control and those
treatments where growth was not obviously affected by ammonia concentra-
tion. The second was based on all remaining treatments plus the results from
the treatment with the highest ammonia concentrations from the first regres-
sion analysis. The point of intersection estimates the maximum concentra-
tion of ammonia that did not reduce prawn growth. This point and the second
regression equation were used to predict the concentration of ammonia ex-
pected to result in a 5% reduction in growth (Figs. 1 and 2).

RESULTS

Experiments 1 and 2 -Acute bioassays

Mortality of school prawns over 96 h increased from 33.3% to 100% as
ammonia increased from 1.33 to 2.64 mg NH,,-N/l (Table 1). Following the
272 G.L. ALLAN ET AL.

TABLE 1

Mortality of school prawns (Metupenaeus mackayi) exposed to a range of ammonia concentrations

(Experiment I)’

Ammonia (mg N/l)2.3 Mortality (%)

Total Un-ionised 48 h 96 h

0.06kO.01 0.003 k 0.0003 3.3 i 3.3 lO.Oi 10.0

25.0 2 0.3 1.33rto.02 16.7k 12.0 33.3 * 14.5
31.0rto.7 1.60+0.06 43.32 16.7 90.0+ 5.8
41.2* 1.2 2.19kO.06 86.7 t 6.6 96.7+ 3.3
49.6 + 2.4 2.64kO.13 100.0 100.0
66.45 1.7 3.5410.09 100.0 100.0
‘Values are means + s.e. (n = 3) for each treatment.
‘Mean of the average values (based on daily measurements) for each replicate aquarium.
‘Average water temperature, pH and salinity values for all treatments were 25.1 “C. 8.0 k 0.1 and
34.5%0. in that order.

TABLE 2

Mortality of leader prawns (Penaeus monodon) exposed to a range of ammonia concentrations (Ex-
periment 2)’

Ammonia (mg N/l)2.3 Mortality (O/k)

Total Un-ionised 48 h 96 h

0.04 + 0.003 0.002 + 0.0003 0 0

10.5+0.3 0.48f0.01 0 0
21.2kO.7 0.96 f 0.02 0 0
3l.OkO.4 1.42kO.05 0 3.3k3.3
41.1-tl.l 1.85kO.03 10.0k5.8 86.7k3.3
5l.Ok2.4 2.27 + 0.05 43.3k8.8 96.7 + 3.3

‘Values are means + se. (n = 3) for each treatment.

‘Mean of the average values (based on daily measurements) for each replicate aquarium.
jAverage water temperature. pH and salinity values for all treatments were 26.0fO.l “C, 8.0 and
34.0°/oo,in that order.

high mortalities in four of the six treatments (90.0- 100%)) lower concentra-
tions were used in the acute bioassay with leader prawns (Table 2). With
leader prawns low mortalities (O-3.3%) were recorded up to 1.42 mg NH3-
N/l with mortality increasing to 96.7% as the concentration increased to 2.27
mg NH,-N/l (Table 2). Lethal concentrations (96-h LC5,, values) were higher
for leader prawns ( 1.69 mg NH,-N/l, 95% confidence limits; 1.69, 1.84 mg
NH3N/1) than for school prawns ( 1.39 mg NH,-N/l, 95% confidence limits;
1.22, 1.57 mg NH,-N/l).
TOXICITY OF AMMONIA TO PRAWNS 273

TABLE 3

Mortality of leader prawns (Penaeus monodon) at different ammonia and dissolved-oxygen concen-
trations (Experiment 3)’

Treatment Ammonia (mgN/l)2,3 Dissolved’ Mortality3,J

oxygen (f&l ) (O/o);96 h
Total Un-ionised

Not aerated
I 0.1 kO.01 0.004 k 0.0004 2.2 + 0.09 6.7 ?I 6.7ab
2 12.6k0.2 0.53kO.04 2.2 ?I 0.09 0”
3 19.9kO.3 0.95~0.01 2.2 + 0.03 3.3*3.3ab
4 27.0f0.4 1.3OkO.02 2.2 k 0.06 lO.Ok 10.0ab
5 33.9kO.5 1.63kO.03 2.3 ?I 0.03 90.02 5.8’

Aerated
6 33.5 50.5 1.6OkO.03 5.7kO.01 33.3 !I 5.P

‘Values are means f s.e. (n = 3 ) for each treatment.

‘Mean of the average values (based on daily measurements) for each replicate aquarium.
3Average water temperature, pH and salinity values for all treatments were 26.0 k 0.2 “C, 8.0 I! 0. I and
3 I .O”/~, in that order.
4Means, within a column, sharing a common superscript are not significantly different (P> 0.05 ).
Data were transformed (arcsin .I?.‘) prior to statistical analysis.

TABLE 4

Effects of ammonia on weight gain, survival and food conversion ratios for school prawns (Metupen-
arus macleayi) over 2 1 days (Experiment 4) ’

Ammonia (mg N/1)2.3 Wt. gain4 Survival4 FCR“’

(g/ (Oh)
Total Un-ionised prawn )

0.3f0.01 0.01 kO.0 1.5kO.l” 100” 2.4 k 0.26”

2.2f0.04 0.10+0.003 1.6kO.2” 100” 2.2+0.57”
6.2kO.l 0.27 k 0.003 1.6kO.2” 100” 2.1 kO.20
13.6tO.l 0.64 5 0.09 1.3kO.2” 90.0+0.6” 2.8 t 0.72”
18.5 k 0.2 0.89kO.01 0.6 k 0.2b 73.3 k 1.Zab *
24.2 k 0.6 1.2OkO.04 0.5k0.1b 50.0* 1.2b *

‘Values are means k se. (n = 3 ) for each treatment.

‘Mean of the average values (based on daily measurements) for each replicate aquarium.
‘Average water temperature, pH and salinity values for all treatments were 25.0 k 0.2”C, 8.0 k 0.1 and
3 1.5O& in that order.
4Means, within a column, sharing a common superscript are not significantly different (P> 0.05 ).
‘Data were transformed (log x) prior to statistical analysis.
*Excessive mortality ( > 80%) for reliable FCR estimation.

Experiment 3 - Effect of low DO on acute ammonia toxicity

At a low DO level (2.2 mg/l) mortality of leader prawns was low (0- 10.0%)
and not significantly affected (P> 0.05 ) by ammonia concentration in the
274 G.L. ALLAN ET AL.

TABLE 5

Effects of ammonia on weight gain, survival and food conversion ratios for leader prawns (Penaus
r~ordon ) over 2 1 days ( Experiment 5 ) ’

Ammonia (mg N/I )2.3 Wt. gain4 Survival’ FCR“,’

(g/prawn 1 (Oh)
Total Un-ionised

0.30*0.03 0.02 + 0.00’ 3.8 k 0.26” 90.0? 1.00 1.6kO.17”

2.1 kO.04 0.10~0.01 3.8 k 0.27” 100 1.4kO.06”
9.8k0.3 0.45 k 0.02 2.9kO.18” 96.1 kO.33” 1.7~0.07”
17.1 kO.2 0.78*o.ol 1.3k0.20b 100” 2.6 -t 0.29b
24.9 k 0. I I .08 * 0.02 0.4 + 0.06’ 46.7 k 0.88” *
32.9+ 1.0 1.60+0.07 _ 0’ *

‘Values are meansIs.e. (n=3) for each treatment.

‘Mean of the average values (based on daily measurements) for each replicate aquarium.
‘Average water temperature, pH and salinity values for all treatments were 26.9 f 0. I “C, 8.0 + 0. I and
36.0%0. in that order.
4Means, within a column. sharing a common superscript are not significantly different (I’> 0.05).
‘Data were transformed (log X) prior to statistical analysis.
*Excessive mortality ( > 80%) for reliable FCR estimation.

range 0.004-l .30 mg NH,-N/I (Treatments 1-4; Table 3 ). However, at 1.63

mg NH,-N/l, 90.0% died at the low DO level. At a similar ammonia level
( 1.60 mg NH,-N/l) prawns held in well-oxygenated water (5.7 mg/l) had
significantly (PcO.05) lower mortality (33.3%).

Experiment 4 - Chronic bioassay with school prawns

Female prawns, in the presence of males, grew faster (P-=0.05) than males
in all treatments. As there was no sex-ammonia interaction (P> 0.05 ) the
data for both sexes were combined to compare means (Table 4) and to esti-
mate the ECS (Fig. 1). Growth relative to the control (0.01 mg NH,-N/l)
was not significantly (P> 0.05) reduced from 0.1 O-O.64 mg NH,-N/l; how-
ever, at 0.89 and 1.20 mg significant reductions occurred (PcO.05) (Table
4). The EC5 value was 0.35 mg NH,-N/l (7.6 mg total ammonia-N/l). Sur-
vival was high (90- 100%) at 0.01-0.64 NH,-N/l but was significantly re-
duced at the highest concentration ( 1.20 mg NH,-N/l) (PC 0.05). Food con-
version ratio (FCR) was unaffected at ammonia levels where survival
exceeded 80% in each replicate (P> 0.05) (Table 4).

Experiment 5 - Chronic bioassay with leader prawns

Growth relative to the control (0.02 mg NH,-N/l) was not significantly
reduced (P> 0.05) from 0.10-0.45 mg NH,-N/l; however, at 0.78 and 1.08
mg significant reductions occurred (P-c 0.05) (Table 5 ). The EC5 value was
0.2 1 mg NH,-N/l (4.1 mg total ammonia-N/l). Ammonia concentration also
TOXICITYOFAMMONlATOPRAWNS 275

significantly reduced survival (PcO.05) at 1.08 and 1.60 mg NH3-N/l and

FCR (PC 0.05 ) at 0.78 mg NH,-N/l.

DISCUSSION

During all experiments, survival among controls exposed to negligible am-

monia levels was > 90%, and relatively stable ammonia, pH, temperature and
DO levels were maintained in all aquaria. Although stocking density in the
aquaria was equivalent to approximately 45 prawns/m’, growth of school
prawns in the controls was 0.07 g/day during Experiment 4. This compared
well with that recorded during four 8-12-week farming trials (0.06-0.08 g/
day) with this species at between 17.4 and 27.0 prawns/m’ in a l-ha pond
(Maguire and Allan, 1985 ). Similarly, growth of leader prawns (0.18 g/day)
in the controls in Experiment 5 was comparable with that measured by the
authors in model ponds (0.17t0.01 g/day) at 25 prawns/m’ and a similar
average water temperature (269°C). However, faster growth rates for this
species (0.3 g/day) have been recorded (Liao, 1977). As survival and growth
rates were comparable with larger scale facilities, the continuous-flow bioas-
say system used here was considered appropriate for assessing both the acute
and the chronic toxicities of ammonia.
The acute toxicity results obtained are compared with other published val-
ues for crustaceans in Table 6. As different levels of water quality variables
which affect the NH3/NHt equilibrium were used, e.g. pH, comparisons are
made between levels of un-ionised ammonia (NH,-N), the more toxic form
of ammonia. Previous studies with penaeid larvae have shown that tolerance
to acutely toxic levels of ammonia increases with age (Jayasankar and Mu-
thu, 1983; Chin and Chen, 1987). For example, the 24-h L&, for leader
prawns increased from 0.54 to 4.70 mg NH,-N/l as the animals progressed
from nauplii to postlarvae (Chin and Chen, 1987 ). Here the 96-h LCsO value
for juvenile leader prawns ( 1.69 mg NH,-N/l) was higher than for postlarvae
( 1.04 mg NH,-N/l) (Chin and Chen, 1987) although the 48-h LCsO values
were more similar (2.33 and 2.50 mg NH,-N/l, respectively).
Wickins ( 1976) used small (0.5-l .5 g ) juvenile penaeids of seven mixed
species and found a lower 48-h LCsOvalue ( 1.29 mg NH,-N/l ) than those for
school and leader prawns determined here. Differences in size (age) and spe-
cies may have accounted for differences in susceptibility. The 48-h LCsO val-
ues for un-ionised ammonia for leader prawns (present study; Chin and Chen,
1987) are the highest equivalent values presented in Table 6 indicating that
this species is somewhat more tolerant, than related species, of acutely toxic
levels of ammonia.
In a chronic bioassay for low DO, Seidman and Lawrence ( 1985 ) found
that while growth of leader prawns was depressed at a very low DO level ( 1.2
mg/l), neither survival nor growth were affected at a level of 2.2 mg/l or
276 G.L. ALLAN ET AL.

TABLE 6

Comparison of acute ammonia toxicity to aquatic crustaceans

Species Stage LCSO Un-ionised ammonia

(h) (mg NH3-N/l)

Homarus’ 4th~stage Incipient’ 1.40

americanus larvae
Macrohrachiurn’ larvae 3-8 days 144 0.26”
rosenbergii post-hatch 0.80b
1.35’
Penaeid prawn? juveniles 48 1.29
(Data for seven (0.5-1.5 g)
species pooled)
Penaeu? nauplii 24 0.29
indrcus protozoeae 24 0.95
mysis 24 3.17
protozoeae 48 1.18
nauplii-postlarvae Incipient’ 0.93
Penaeu? nauplii 24 0.54
monodon protozoeae 24 0.16
mysis 24 2.11
mysis 48 I .30
postlarvae 24 4.70
postlarvae 48 2.50
postlarvae 96 I .04
Merapenaeus6 juveniles 48 1.66
rnacleayi (0.9-4.8 g) 96 1.39
Penae& juveniles 48 2.33
rnonodon (0.8-4.5 g) 96 1.69

‘Delistraty et al. ( 1977).

‘Armstrong et al. ( 1978); “pH=6.83, “pHz7.60, ‘pH = 8.34.
‘Wickins ( 1976).
“Jayasankar and Muthu ( 1983 j.
5Chin and Chen (1987).
‘Present study.
‘Incipient LCse - lethal concentration for 50% of individual on long exposure.

above. In the present study, a DO level of 2.2 mg/l together with ammonia
levels ranging from 0.004-1.30 mg NH,-N/l (Table 3) had no significant
effect on survival of leader prawns. However, at higher ammonia levels ( 1.60-
1.63 mg NH,-N/l) mortality was much greater (90%) at a DO level at 2.3
mg/l than at a DO level of 5.7 mg/l (33.3%). This demonstrates that low DO
increases acute ammonia toxicity. In Experiment 3 the 96-h LCSOat low DO
levels (2.2-2.3 mg/l) lay in the range 1.30-l .63 mg NH,-N/l, compared with
the 96-h LCSO,estimated in well-oxygenated water (Experiment 2), or of 1.69
mg NH,-N/l (95% confidence limits 1S6, 1.84 mg NH,-N/l). This indicates
that the low DO caused a reduction in the acutely lethal level of ammonia to
leader prawns of up to 0.5 mg NH,-N/l. For this comparison to be valid the
TOXICITY OF AMMONIA TO PRAWNS 211

two data sets (Experiments 2 and 3) must be comparable. To confirm this

assumption, the mortality data (33.3%) for Treatment 6 (high DO; 1.60 mg
NH,-N/l) in Experiment 3 (Table 3) was used in the probit regression equa-
tion generated from the acute bioassay in oxygenated water (Experiment 2 ).
The ammonia concentration which was predicted to give this mortality after
96 h was 1.6 1 mg NH,-N/l (95% confidence limits 1.47, 1.76 mg NH,-N/l).
This compares very well with the actual ammonia concentration used for
Treatment 6 in Experiment 3 ( 1.60 mg NH,N/l) (Table 3 ).
Toxicity studies of water quality variables to aquaculture species usually
indicate “safe” levels for routine farm management. Here ECS values, calcu-
lated from 3-week experiments, were used to indicate the maximum accept-
able levels, For school and leader prawns these were 0.35 and 0.2 1 mg NH3-
N/l, respectively. In the only other published study on the effects of ammonia
on penaeid growth, the average of the EC1 and EC2 values (those concentra-
tions which reduced growth by 1 and 2%) was used as an estimate of the
“maximum acceptable” level (Wickins, 1976 ) . Six mixed species were used
and the level reported was 0.10 mg NH,-N/l (Wickins, 1976). If the same
technique is applied here the “maximum acceptable” level for school and
leader prawns would be 0.30 and 0.17 mg NH,-N/l respectively, higher than
Wickins’ value and indicating that school and leader prawns are also tolerant
of sub-lethal concentrations of ammonia.
The higher ECS value estimated for school prawns compared with leader
prawns was anomalous, considering that the 96-h L&, for school prawns was
lower ( 1.39 mg NH,-N/l) than that for leader prawns ( 1.69 mg NH,-N/l).
The two species may differ in their relative susceptibility to acute and chronic
ammonia poisoning. However, it should be noted that during the chronic
bioassay with leader prawns prolonged dry weather resulted in elevated sal-
inities (36.0%0) which were above the optimum for osmoregulation (Caw-
thorne et al., 1983) and this may have affected their susceptibility to chronic
ammonia poisoning. Increased salinities have been shown to reduce ammo-
nia excretion rates (Spaargaren et al., 1982; Claybrook, 1983). In contrast,
salinities during the chronic bioassay with school prawns were lower ( 3 1.5%)
and close to their optimum for osmoregulation (Maguire and Allan, unpub-
lished data, 1989).
The concentration of ammonia in prawn ponds or tanks should not be al-
lowed to rise above “maximum acceptable” levels (EC,) and should cer-
tainly not be allowed to approach lethal levels (96-h LCsO). Total ammonia
levels in excess of the LCso values determined here have been recorded during
intensive culture of Penaeus penicihtus (Chen et al., 1988); however, spe-
cies tolerances do differ. Temperature, salinity, dissolved oxygen and pH
should also be monitored as unfavourable levels may synergistically affect
ammonia toxicity.
278 G.L. ALLAN ET AL.

ACKNOWLEDGEMENTS

The authors would like to thank Mr. Arthur Woods (University of N.S.W.)
for his assistance with statistical analyses and Mr. S. Battaglene, Dr. J.A. Nell,
Dr. R.J. MacIntyre and two anonymous referees for their constructive com-
ments on the manuscript. Research grants from the Fishing Industry Re-
search Development Council (FIRDC) of Australia and the Reserve Bank of
Australia (Rural Credits Development Fund) are gratefully acknowledged.

REFERENCES

Alabaster, J.S., Shurben, D.G. and Knowles, G., 1979. The effect of dissolved oxygen and sal-
inity on the toxicity of ammonia to smelts of salmon, Sultno salar L. J. Fish Biol., 15: 705-
712.
American Public Health Association (A.P.H.A. ), 197 1. Standard Methods, 13th edn. A.P.H.A..
Washington DC, 874 pp.
Armstrong, D.A., Chippendale, D., Knight, A.W. and Colt, J.E., 1978. Interaction of ionized
and unionized ammonia on short-term survival and growth of prawn larvae, Macrobruch-
iutn rosenbergii. Biol. Bull., 154: 15-3 1.
Boyd, C.E., 1982. Water Quality Management for Pond Fish Culture. Elsevier, Amsterdam,
318 pp.
Bower. C.E. and Bidwell, J.P., 1978. Ionization of ammonia in seawater: effects of temperature,
pH and salinity. J. Fish. Res. Board Can., 35: 1012-1016.
Busvine, J.R., 1957. A Critical Review of the Techniques for Testing Insecticides. Common-
wealth Institute of Entomology, London, 208 pp.
Cawthome, D.F., Beard, T., Davenport, J. and Wickins, J.F., 1983. Responsesofjuvenile Penaeus
monodon Fabricius to natural and artificial sea waters of low salinity. Aquaculture, 32: 165-
174.
Chen, J.C., Liu, P.C. and Lin, Y.T., 1988. Super intensive culture of red-tailed shrimp Penaeus
penicillutus. J. World Aquacult. Sot., 19 (3 ): 127- I3 1.
Chin, T.S. and Chen, J.C., 1987. Acute toxicity of ammonia to larvae of the tiger prawn, Penaeus
tnonodon. Aquaculture, 66: 247-253.
Claybrook, D.L., 1983. Nitrogen metabolism. In: L.H. Mantel (Editor), The Biology of Crus-
tacea. vol. 5 - Internal Anatomy and Physiology Regulation. Academic Press, New York,
NY, pp. 163-213.
Colt, J.E. and Armstrong, D.A., 1981. Nitrogen toxicity to crustaceans, fish and molluscs. In:
L.J. Allen and E.C. Kinney (Editors), Proceedings of the Bio-Engineering Symposium for
Fish Culture. Fish Culture Section of the American Fisheries Society (FCS pub]. 1)) Bethesda,
MD, pp. 34-47.
Dal Pont, G., Hogan, M. and Newell, B., 1973. Laboratory Techniques in Marine Chemistry.
II. Determination of Ammonia in Seawater and the Preservation of Samples for Nitrate
Analysis, Aust. C.S.I.R.O. Div. Fish. Oceanogr. Rep., C.S.I.R.O., Sydney, 11 pp.
Delistraty, D.A., Carlberg, J.M., Van Olst, J.C. and Ford, R.F., 1977. Ammonia toxicity in cul-
tured larvae of the American lobster (Hotnarus americanus). Proc. Annu. Meet. World
Maricult. Sot., 8: 647-672.
Jayasankar, P. and Muthu, M.S., 1983. Toxicity of ammonia to the larvae of Penaeus indicus
H. Milne Edwards. Indian J. Fish.. 30( 1): l-l 2.
TOXICITY OF AMMONIA TO PRAWNS 279

Klima, E.F., 1965. Evaluation of Biological Stains. Ink and Fluorescent Pigments as Marks for
Shrimp. U.S. Fish Wildl. Serv. Spec. Sci. Rep., No. 51 I, 8 pp.
Liao, l.c., 1977. A culture study on grass prawn, Penaeus monodon, in Taiwan - the patterns,
the problems and the prospects. J. Fish. Sot. Taiwan, 5( 2): 1l-29.
Lloyd, R., 196 1. Effect of dissolved oxygen concentrations on the toxicity of several poisons to
rainbow trout (Salmo gairdneri Richardson). J. Exp. Biol., 38: 447-455.
Maguire, G.B. and Hume, I.D., 1982. A study of the nutritional requirements of school prawns
Metapenaeus marleayi (Haswell) in some Australian brackish water farming ponds. Aqua-
culture, 29: 26 I-278.
Maguire. G.B. and Allan. G., 1985. Development of methods for growing juvenile school prawns,
Metapenaeus macleayi. in estuarine ponds. In: P.C. Rothlisberg, B.J. Hill and D.J. Staples
(Editors). Second Aust. Natl. Prawn Sem., _32-26 October 1984. NPSZ. Cleveland, Aus-
tralia, pp. 337-351.
Maguire, G.B., Allan, G.L.. Baigent. R. and Frances, J., 1988. Evaluation of the suitability of
some Australian and Taiwanese diets fed to leader prawns (Penaeus monodon) in ponds. In:
L.H. Evans and D. O’Sullivan (Editors), Proc. First Aust. Shellfish Aquacult. Conf., Perth,
Australia, 23-27 October 1988. Curtin University of Technology, Perth, pp. 89-l 04.
Major, G.A., Dal Pont, J., Kyle, J. and Newell, B., 1972. Laboratory Techniques in Marine
Chemistry. A Manual. Aust. C.S.I.R.O. Div. Fish. Oceanogr. Rep., No. 5 1, C.S.I.R.O., Syd-
ney. 55 PP.
Mount. D.I., I96 1. Development of a system for controlling dissolved-oxygen content of water.
Trans. Am. Fish. Sot., 90(3): 323-327.
Mount. D.I., 1964. Additional information on a system for controlling the dissolved oxygen
content of water. Trans. Am. Fish. Sot., 93: 100-I 03.
Scribner, E.A., 1986. Report on seawater supply, Brackish Water Fish Culture Research Station.
Div. Fish., N.S.W. Dept. Agric. Int. Rep. (unpubl. ). 6 pp.
Sedgwick, R.W.. 1979. Influence of dietary protein and energy on growth. food consumption,
and food conversion efficiency in Penaeus merguiensis De Man. Aquaculture. 16: 7-30.
Seidman, E.R. and Lawrence, A.L., 1985. Growth, feed digestibility, and proximate body com-
position of juvenile Penaeus vannamei and Penaeus monodon grown at different dissolved
oxygen levels. J. World Maricult. Sot., 16: 333-346.
Shaw. J., 1960. The absorption of sodium ions by the crayfish Astacuspallipes Lereboullet. III.
The effect of other cations in the external solution. J. Exp. Biol., 37: 548-556.
Smart, G.R.. 1978. Investigations of the toxic mechanisms of ammonia to fish - gas exchange
in rainbow trout (.Sahro gairdneri) exposed to acutely lethal concentrations. J. Fish Biol..
12: 93-104.
Sokal, R.R. and Rohlf, F.J., 1981. Biometry. W.H. Freeman, San Francisco, CA, 859 pp.
Spaargaren, D.H., Richard, P. and Ceccaldi, H.J., 1982. Excretion of nitrogenous products by
fenaeusjaponicus Bate in relation to environmental osmotic conditions. Comp. Biochem.
Physiol., 73A(4): 673-678.
Sprague, J.B., 1969. Measurement of pollutant toxicity to fish. I. Bioassay methods for acute
toxicity. Water Res., 3: 793-82 1.
Stanier. R.Y., Aderlberg, E.A. and Ingraham, J.L.. 1976. General Microbiology. Macmillan,
London, 87 1 pp.
Tarazona. J.V., Munoz. M.J., Ortiz, J.A., Nunez, M.O. and Camargo, J.A., 1987. Fish mortality
due to acute ammonia exposure. Aquacult. Fish. Manage., 18: 167- 172.
Thurston. R.V., Phillips, G.R. and Russo, R.C.. 198 1. Increased toxicity of ammonia to rain-
bow trout (Saimo gairdneri) resulting from reduced concentrations of dissolved oxygen.
Can. J. Fish. Aquat. Sci., 38: 983-988.
Whitfield, M., 1974. The hydrolysis of ammonium ions in sea water - a theoretical study. J. Mar.
Biol. Assoc. U.K., 54: 565-580.
280 G.L. ALLAN ET AL.

Wickins, J.F.. 1976. The tolerance ofwarm-water prawns to recirculated water. Aquaculture, 9:
19-37.
Winer, B.J., 197 1. Statistical Principles in Experimental Design, 2nd edn. McGraw-Hill, New
York, NY, 907 pp.