This paper not to be cited without prior reference to the authors
International Council for
the Exploration of the Sea
C.M. 1971/E:8
FisheriesImprovement Committec
Ref.:C(Hydro. Cttee)
Tho uso of a continuous flow apparatus in the study
of longer-term toxicity of heavy metals
by
K. W. Wilson and P. M. Connor
Fisheries Laboratory, Burnham-on-Crouch, Essex
suMMARy
"
•
A continuous flow apparatus has been built at the Fisheries Laboratory
~.
.,
at Burnham-on-Crouch that is capable of administering different, yet constant,
concenfrations of toxin t'o twenty experimental tanks.
Using thisapparatus,
the toxicity of mercury (as mercuric chloride)and cadmium (as cadmium
chloride) to the brown shrimp (Crangon crangon) has been studied over
several wecks.
After 1500 hours' cxposure these metals were toxic at 1/1000
and 1/100 respectively of their 48 hr LC
• No lcthal thrcshold concentra50
Although the toxins did not affect the rates at which
, tion was established.
the animals moulted, at the higher concentrations newly-moulted animals w'ere
rouch more vulnerable.
moulting.
This was shown an
0.
decrease in survival time after
A correlation betwe€n thc size of the shrimp und concentration
was established, with smaller shrimps being more sensitive.
IN1RODUCTION
The techniques of acute toxicity testing in sea water have followed
~
those developed by the freshwater toxicologists, where thc determination cf
the concentration of
0.
toxin
~equircd
to kill' 50 per cant of the test
organisms within 48 ,or 96 hours (48or 96 hour LC
practice.
) is 0. well established
50
However, it is becoming increasingly recognized that these
figures give little or no indication of the pattci-~ of intoxication,and
the' I safe,i concentration of two chomi~als with'tri!e: '~ame 48 hour LC
can
50
differ markedly' (Alabaster 1970). A more importan't 'parameter is tho maximum
. ..
.'
"
~
:~.
...
concentration at which acute toxicity is not ..apparent.
This concentration,
,........
: .
.
"
,-
also known as the incipien~ LC
or lethal threshold conccntratio~" is n?t
50
necessarily the safe conc~tration at which 0. fishory would.survive, for it
takes'z{o a~~o~t of sub-lethal effecte but is, as'Sp~ague:(1969) po~t~ ~~t,
11
~~
.'.
~'c:onvenicnt and reproducible ref~rcncG point:. that co~centr~ti~~'
';;~dd Rill thc averägc. fish on long oxposur~l":" ~~e w~t on to p~int
whlch
,
' . . ' , .. ',:
': ",; ':'
1.,
'
out that in most cases this thrcohold'of acutc toxicityis reachcd within
1
,
. ,.
thi~ VlO.S ~ot:the :E..~9 for~9.~~~~ Cor~.~r..an~LR..~g~er_.. (1.9.5~1:.: .__'... ~.
demonstrated that, in thc statie eonditions of thcir test, moreury was lost
4 dnys;
froma solution of mereurio chloride in sea 'wator by evaporation, by . adsorption totho glass walls of thc aquarium, and by absorption by baeteria in
the.water.
Boetius (1960) dcmonstrated ·the uptakc'of the same metal from
solution by his test fish.
These lossos lead to a docren:se in the dosired
concentration which becomes signifieant at low conccntrations end results
,
.
in an underestimation of thc toxieity of the solution.
The prcsent paper describos an apparatus designcd and built to deliver
continuously low concentrations of toxin at such a rate that losses ta.the
.....
.tank und animals had negligible effccts on thc external eoncentration.
i~
. .::;
:.!~
... ,.;
The
~.~
.. apparatus is simple and reliable and cnables teGts to be carried out for
,
.~ ~
.,
..... - i..--:.
long periods with a minimum amount of maintcnanco.
.,
~.
"
".,i:'"
.; ....
.
~
_·:::,;.:.-11:·~~~'·.
d,
Sor:lC prcliminary results
L ...=; '!.,,;;'_.:"
I
;:' .. ,
achieved with the apparatus, using salts of cadmium and mercury, .ar.e
:
';.:
described.
-_
'---'
. :METHODS
!.
:-,
. . r;
........
. • . . : .....
I
•
.,\",! ....
'i'
Thc continuous flow appara~us will be deseribed in detail elsewhcre.,:
:..cConnor and WHson, in pr~paration) so that only a bri.ef description',ish.i;·
\
" given here.
The apparatus:.was'built.to supply eaeh of .twcnty10-litr.~
o\~
'Perspex' treatment aquario.;with its
at a constant coneentratian.
'"
flow of sea watercontaining toxin
Settled seo. water from thelaboratory.supply:
is fil tered and entern ther.(:w~r'\[oir tank v:here i t is heated) (ar: cool cd) ,;to
15°C (see Figure 1).
Water in!tho tank is pumped continuouslyto the
. ;'
header tank at a rate whieh excceds tho demand of thc aquario., thc cxcess
.... :.:_.
.'
' .. :. i.:.... "
s.ea wa~er,. overi}-owing back i~ti.:~ "'~~~ reservoi! tank:. The ~(:ader :~k f~?ds
the treatment nquc.ria individually, eaeh separate flow passing through i ts
. . :
.
.
.,..1.,\· .. ·
,....
flov~eter
own adjustable and calibratcd
: i; -,'
-\f'
....
,.
.: ..
into thc mixing chamber.
'.
'
,
r
' .
'1
:.:.,
Toxin
'.
•
',.'
1.
.'
held in the storage bottle is netercd into the mixing chamber o.t a eonstant
::
rate by
0.
.:.·;;~l'1.~",·f·_~:"-:·:
.
peristo.ltic pump.
] : • •
_,
. " : .:
• .J
., ~ .
I'i
" .
. " .
•
'j.~;i -:.
w~~e::?"and
Here it nixes with the flowing sea
.'':'
. ,.
l..
\,'
~
_
the solution thus formed leaves tbe chambcr und flows into the treatment
.
' ~~. i
'1
~.
aquarium before running to waste.
•
':,
)
t ...
'c-'
f
-'. '
!.'
All flows are adjusto.blc, but in the
;r~~~texpcriments se~.'~~t'e~'tw~s supplied' to e~c'h' ~~~~i~ ~t 10\/h'~~l ; .
•
":
• '\
.~ t .!
,1
i } .
';'
.
.
'.'
.'
. ,'~:-
:" -
• ': .
f' .
.
" " ; ' ! " '.
.••.
.
~d toxin at 6.67 ml/hour;' thc', desired ~oncentrations in tho aquD.ria"~ere
,_
.
. : '
i .
.'
•
.
.
_.
;
~
achieved by' moldng up st~~k sol~tions ~ftoxin at conce~tr~tions1500'ti'~~s
~cater thnn: that 'de~ir~d.:· Stock solutions' ~ere m~de up' every 48'hour~:~
. d~sti'ii~d vlD.tcr ~si~g "~rwiar" ~D.dmium chloride (caC1 2-~H20) ~d me;c~ic
2
.
,",_..
.
.
.
. . ..
chlo~ide (HgCl2).
'.
.:),
'.
in each aqu6.rium but tb~y' ~~c~:{ individually retaincd in smo.ll, extruded
. .
'
Fift.ee~:l1~~~ brovm shrimps (Cran/mn er~gon) were he~d
.' ~ •.
j" J :-"'.l
2
'plastic (IN'etlon I) cages.
day on chopped'nussel.
Tho aniD.als were fed to satiatioh "every third"
The aquaria were
L~npected
animals nnd'maulted cuticles were removed.
mals was measured
~ith
twice daily und all dead
Tho carapace lcngth of 0.11 'tuli-
carlipcrs at death.
The experiments run for two
months.
RESULTS
The distribution of the survival tiDen of the shrimps at each concentration was fOillld to be log normal (typical results for cadmium are given
ihFigure 2);
the resul ts were ano.1ysed gTaphically to determine the' mean
'survival times (ET
SO ) und 9S per cent confidenco liDits(Litchfiold 1949).
The resul ting survival-concentration curvos for mercury and'cddtnium and
•
the results of 48 hour LO
Figuro 3.
SO determinations (Portmann1970) are shown in
In both cases thera is
0.
similar curvilinear relationship between
survival ti~e und concontration, but the incipient threshold concentration,
indicated by tho curve becoming parallel with the time axis, was not reached
after 1500hoursl (62 days~ exposuro. Mortality occurred evcn after continuing exposure at low concentrations.
This is indicative of tho
accumulativc nature of the poisons, and it seoms likoly that the threshold
concentration woUld only become apparent Vihen the effects of uptake of the
toxin are balnncüd by' excration or other ITlethods cf detoxificntion.
.'
By retaining tho shrimps in separate cages, i t has been possibIri: to'"
"stuQy some of tho biological factors affecting theirsusceptibilityto
poisoning.
•
Newly-moulted nnimals were found to be nore'vulnerablo; 'for
examplc, of the 15 aninals that had obulted after 297 hoursl exposure to'
0.1 ppm cadmium, 11 had died 1 wheroas only one of the 15 unmoulted shrimps
had died.
The differcnce \7aS significant (X
2
=
8.32;
P
=
0.01-0.001).
·Thodistribution of the tiDen of Doult waS found to be log normal
(~igure 4), and the significance tests of Litchfield (1949) sho~ed thom to
.'~ ; .. I
" :'
• ,;
I
"
.1:?e ·the same for tho controls nnd' for Shrir.:1PS exposed to all concentrations
/
of. toxins.
".,
Hovlever, post-noult sur-vival
concentration of notal (Figure 5).
~as
inversely related, to,
tho,:, , .
.
~
('"
This increased susceptibility nay be
duo to increased permeability to the toxin following mÜult, to 'the
i~crcased. physiolo~ica~ .stresses a!3sociatod
witl:l moul ting, ,or ,to a combina-
tion of '. thesö factors.
The,~effect
of carapace
leng~h
and 'thcreforo body sizo on the suscepti-
bili ty of shrinps to thetoxins '-is sho'l'.'11 in Figuro 6.
Fron thc survival
. distribution, the population was divided into 20 per 'cent inte~vais
(see
inset to Figure 6), and the average 'carapaco lcngth of tho aniDals dying
.....
• •
3
I
"
-
.,
in each interval was calculated.
To enable conpurison of' lengths und sur-
vival tines f'rom dif'ferentconcentrations the average length of each 20 per
,. cent intervo.l was expressed as a ,percentage of the average curapace length
of' the total distribution at that conccntration.
The results shovm in
Figure 6 ure in agreement with those of Portmnnn (1968), with smaller
shrimps being much more sensitive •
.CONCLUSIONS
The continuous flow apparatus has provcd reliable und accurate during
preliminury tests lasting more thun two months, und from these longer-term
,studies it has .been possible
recognizc toxins that probably excrt their
~o
influence by becoming slowly accumulated.
Analyses of, thc tissucs of dcad
unimo.ls for the metals ure plnnned shortly to confirm this possibility.
These und other tests suggest :that considerablo caution is needed in predicting the cffects of' long-term exposure to toxins on the basis of their
48 hour LC
50
values, unless a ?lenr lethaI threshold concentration has'
bcen demonstrated.
In the tests
descr~bcd,
concentrations of mercury und
cadmium at 1/1000 und 1/100 of the 148 hour LC
50 values respectively htl.vc.,
been shown to belethal in the longer term.
Sources of variation in toxicity tents ure as often derived from
between test populations as from differences in test proThe present paper describes the diff'erence of'bo~ sizeund
heterog~neity
cedures.
moulting" but other biologieal: variables - such as age, condition und sex ure likely to be significunt. The inportuncoof' these f'actors iß not alwnys
evident in short-tcrm toxicity, tests.
REFERENC~S
ALABASTER, J. S., 1970. Testing the toxicity of effluents tofish. ·Chernlf.
lud. 1970, 759-764.
BOETIUS, J., 1960. Lcthal action of nercuric chloride und phenylmercuric
.
acetate on fishes •. Mcddr
Danm. Fisk. -og Havunders., 3(4),93-115.
,
-
CORNER, E. D. S. 'wd RIGLER, F~ H. R. ~ 1958. Tbc nodes of' action of' toxic
,agents. 111. Mercuric:chloride und N-Urnlflmercuric chloride'on
crustaccuns. J. mur. biol. Ass. U.K., 1I, 85-96.
LITCHFIELD, J. T., 1949. A oethod for rapid graphic solution of' time-per
cent effect curves. J. Phurmac.exp~ Thor., 21, 399-408.
PORTMANN, J. E., 1968. Progress report on a progranno of insecticidc
analysis und toxicity-testing in relation to the marine environment.
IIelgolondür wisse Meeresunters. , 17, 247-256.
.
.""
.
-
.
....
PORTMÄNH, J. E., 1970. The toxicity of 120 substnnces to marineorgnnisms.
-?-,
lVlAFF Shellf'ish Inf'ormati'on Leaflet No. 19. Fishcries Laboratory, .
. .'...:',: Burnham-on-Crouch, Esnox, 10 pp.
'
SPRAGUE, J. B., 1969. Measuroocnt of pollutcnt toxicity to f'ish. 1. Bioass~ methods for acute toxicity.
Water Renenrch, 2, 793-822.
4
•
...
Header tank
•
1
i
_____ ..J
I
I
I
I
,
I
I
I
I
I
I
I
,
I
V
-<-:---
L.--
I
I Perspex
11
•
treatment aquarium
L
.~
Heater
=
~-----
~.
Storage
bottle
_
eooling coil
Overflow [.
Seawater supply'
PumpFigure 1
Reservoir tank
Diagram of the continuous flowapparatus. Only one treatment unit is shown.
"
t t
98
•
3·3 ppm
1·0 ppm
•
90
0
0·32 ppm
•
Q)
"
.-J
.0
U
(/)
E
50
a..
.....>.....c
L-
a
::E
~
0
10
2
0 ' - - - - - - - - - - - - - - - - - - - - - - - - -......
10
Figure 2
100
Survivo l time, (hours)
1000
The accumulative percentage mortaUty curves for the brown shrimp in
different concentrations of cadmium as cadmium chloride. The ET50
is the time that each Une intersects 50 per cent mortality.
,,'
Mercury [Hg++]
1000
1
1
100
II
I
I
'-
10
I
(Portmann 1970)
(f)
~
4BhlC so
L-
L..-
O'Ol
~
..__I
__L_
0·1
1·0
10
Cadmium [Cd++]
1000
100
~.
4BhlC so
10 '"-
'"-
I
....1.-
1·0
(Portmann 1970)
--1-
10
I
---1
100
Concentrotion of' metol (ppm)
Figure 3
Survival-concentration curves for Crangon in mercuric chloride and cadmium
chloride solutions. Mean and 95 per cent confidence limits are shown for each
determination.
'.
•
95
0
()
A
.
·!:J..t A
V A A
VA
A
-
-c
Ql
.
V
:::>
~
0
E
V
Ql
c
c:
vA.
Ql
u
" L-
, Ql
a..
A
A.
0
A
t:::n
,
•
0
0
0
I A •a
•.
A
!:J.
A
O
v
...J
..
6~
\~&
V
50
!:J..
,
~
()
(I
a
C.
0(1
•
(I
,5
Nominal concentrations ppm
0
Cadmium
(I
•
0·10
0·033
Control
v 0·033
Mercury
0·010
•A 0·0033
!:J.
Control
0
10
100
1000
'. Time to first mault otter start of experiment (hours)
..
"
'
Figure 4
moults after the start of the
The time distributions
., of first
,
experimental period.
.,<
.
•
,"
..-
.~1000
~
----0
::;)
- - - - - - --- --- -----
0 _ _ _ _ _ -.: _
E
cu
V)
.......
~
- 100
r:J
CI)
L-
--- 0 .... - o
'\
\
~o
0
L-
0
..,
0
---L-
" "'-
\
\
" ,
" '\
\
'0
~
c:
0
C-l
~
0
'\
..s=
'\
0
cu
E
'.
\
,
.'\
0\
\
'\
......
\
\
r:J
---.~
• Cadmium
>
L~
\
\
\
\
Mercury
0
CI)
\
\
\
\
\
c:
\
c::J
cu
::E
10
I
I
0 I 0'001
Figure 5
120
I
0·01
Concentration of metal
\
0
•
\
oe
',0
0·1
(ppm)
Effect of concentration on time of survival after moulting.
• 0·10
A 0·33
o 1·00
• 3·33
110
I
,
o
ppm
ppm
•
ppm
ppm
cu
o
u
A
c::J
c.
A
~ 100
•
u
c:
cu
E
>-
•
o
c:J
90
~
..-c:J
E 80
E
o
~
~
40
c:J
:;
E
20
:::I
Cl
BO
cu
~
U
L-
cu
0..
60
GI
cu
~
c::
100
Cl
70
L..-
--I-
"""---
0.....:::;----Time
~
0-20
20-40
40-60
' 60-80
Percentage intervals of survival distributions
Figure 6
~
BO-100
Influence of carapace length on the susceptibility of Crangon to mercury
poisoning. • 3. 33; 0 1. 0; A 0.33; • 0.10 ppm of mercury.
Inset ,shows a typical survival distribution for one concentration with the
20 per cent intervals indicated.