&EPA
United States
Environmental Protection
Agency
Environmental Research
Laboratory
Narragansett Rl 02882
Research and Development
EPA-600/S3-82-013 Sept. 1982
Project Summary
•N
S
Sublethal Effects of Number 2
Fuel Oil on Lobster Behavior
and Chemoreception
Jelle Atema, E. B. Karnofsky, S. Olszko-Szuts, and B. Bryant
The experiments described here are
designed to determine the oil exposure
levels at which lobsters show behav-
ioral abnormalities and inappropriate
responses. These levels were deter-
mined as 0.1 to 1.0 parts per million
(ppm) of oil in water. The behavioral
abnormalities can lead to lack of
feeding and subsequent population
decline; they occur at exposure levels
below those levels that cause obvious
loss of equilibrium and coordination
(levels over 1 ppm), eventually leading
to death of the organism.
Extensive control measurements
were incorporated in the experimental
design to ensure that the observed
behavioral changes were due to oil
exposure and not to natural variability.
Rigorous chemical procedures deter-
mined the actual exposure levels in the
lobster tanks. To understand the
consequences of the behavioral ab-
normalities measured in these highly
controlled and, thus, artificial labora-
tory experiments, the results must be
interpreted in the context of the
lobster's natural behavior and ecology.
Such field studies and naturalistic
observations are areas of active
research in this laboratory.
In an attempt to explain the mechan-
isms by which the behavior deterio-
rates, two topics were examined:
interference with normal smell and
taste, and change of motivation. A
combination of neurophysiological
and behavioral experiments on chemo-
reception were designed to investi-
gate these topics.
This Project Summary was developed
by EPA's Environmental Research
Laboratory. Narragansett, Rl, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Research on the effects of oil pollution
on marine organisms is a matter of
general concern because of increasing
tanker transportation, offshore drilling
and the companion risk of oil spills. High
concentrations of petroleum hydrocar-
bons can be lethal to many marine
species. Lower, sublethal concentrations
may interfere with certain life processes
such as mating and reproduction,
feeding and growth, and defense against
predation. Over time, reduced efficiency
in these processes may decrease
populations without directly killing
individuals Sublethal pollution may
shift the ecological balance of affected
areas, resulting in the eventual dis-
appearance of desirable species (those
beneficial to man) and, possibly, the
proliferation of undesirable species.
Such was the case after the well-
studied 1969 West Falmouth, Massa-
chusetts, spill of No. 2 fuel oil (Sanders
et a/.. 1980). The rich fish and lobster
grounds of the North Atlantic may be
directly affected.
One sublethal effect may be that oil
can interfere with the chemical signals
vital to marine life. Most of these
organisms use chemical signals more
-------
than vision and hearing for feeding,
hunting, scavenging, mating, habitat
selection, migration, alarm and escape.
Mounting evidence also indicates that
chemical signals from the egg, in both
plants and animals, attract sperm cells
to it. Petroleum hydrocarbons interfere
with this attraction causing decline of
algal populations. In general, interference
with chemical communication systems
can be expected to have significant
consequences, some obvious, some
not. Thus, animals that show no signs of
locomotor difficulties may still have
sensing problems in feeding, finding
mates, and escaping from predators.
Detecting chemical stimuli has been the
focus of several of our studies on
pollution interference with normal
lobster behavior.
Speculation of petroleum hydrocarbon
interference with chemoreception has
appeared frequently in the literature,
starting with Blumer (1970). The
compounds that animals use for com-
munication and orientation have chem-
ical features in common with compounds
in petroleum, such as carbon skeleton,
functional groups, volatility and solubility.
The chemical look-alikes in oil may
mimic or mask the reception of biologi-
cally important signals. Mimicked
signals may cause "false alarms", with
animals looking for food or mates, or
avoiding predators, where there is none.
Alternatively, if the chemical signals are
masked, animals may miss opportunities
to feed, mate or escape. Another
possibility, less often mentioned, is
that animals may receive two competing
signals, such as an attracting signal
from food and a repelling signal from oil
In this case, while chemoreception
processes may be normal, the animal
would not be able to decide whether to
feed or hide. Any hesitation might be
critical since even slight delays in
responding to food can put the animal at
a disadvantage in competition with an
unimpaired animal.
This study is the first documentation
of the specific way in which oil interferes
with chemoreception, using behavioral
combined with neurophysioiogical
analyses. These studies can provide a
better insight into the potential conse-
quences of oil pollution, as the physio-
logical processes of chemoreception are
probably similar in all animals. Also,
interference with chemoreception or
chemically mediated behavior is one of
the most sensitive biological measures
of low-level oil pollution.
The lobster, Homarus americanus,
was chosen for the study of sublethal
effects of low-level oil pollution for a
number of reasons. The lobster lives in
and on the floor of inshore and conti-
nental shelf waters, areas often affected
by oil spills and chronic discharge. In
some places, the lobster isthedominent
benthic species, so that a decrease in
the lobster population could have
widespread ecological ramifications
The lobster supports an important
commercial and recreational fishery,
and is a symbol of the region itself. Its
decline would have significant socio-
economic consequences. Finally, exten-
sive background data on the lobster
already exists, along with oil and
drilling mud toxicity studies on various
life stages Several of these studies
have been and are being conducted in our
laboratory (Figure 1) which facilitates
interpretation of results from controlled
laboratory experiments
Lobsters use chemical cues from their
environment to direct a number of vital
behavioral responses, such as feeding,
courtship and larval settlement Oil may
alter many of these. For this study we
selected bait localization, which is
crucial to survival, is noticeably affected
by oil, and is amenable to laboratory
testing. In conjunction with behavioral
observations, we studied the effects of
oil on the two major chemoreceptor
systems The antennular system (smell)
is normally used to alert the lobster to
the presence of food and to convey
directional information for odor localiza-
tion The dactyl chemoreceptor system
(taste) is concerned with food evaluation
and feeding This study, using the
water-accommodated fraction (WAF) of
No. 2 fuel oil, considers both distance
chemoreception of the antennules and
dactyls, and contact chemoreception of
dactyls and maxillipeds.
The first purpose of these experiments
is to determine the range of No 2 fuel oil
exposures affecting the feeding behavior
of lobsters without causing neuromus-
cular disturbance. The second goal is to
examine the effects of chemoreceptors
in animals in which sublethal behavioral
abnormalities have been observed. This
approach was used to determine whether
the behavioral abnormalities result
from oil-induced malfunction of the
chemoreceptors, or from what we have
called in other studies lack of "motiva-
tion" (Atema and Stein, 1974).
Toxicity studies such as this one must
be carried out under well-controlled
laboratory conditions to be able to
document the specific effects of given
concentrations and to compare differ-
ent stages in the life cycle and different
seasons. At the same time, the tests
would have little relevance without an
understanding of the animal's behavior
in the natural environment. In order to
develop a complete picture of the impact
of oil pollution on lobster behavior, the
following sequence of experiments
should be undertaken: 1) nature studies
to determine the general context in
which the animal evolved and presently
lives, and to determine its sensitivity to
particular stresses of its environment;
2) laboratory studies to quantify promis-
ing behavioral measures; 3) detailed,
rigorously controlled laboratory tests to
collect a data base to generate a
response model; and 4) field studies on
the same behavioral effects to verify
laboratory results. Field studies on the
American lobster are currently under-
way. Since these studies are technically
and physically difficult, an intermediate
step has been successfully used which
employs naturalistic large aquaria,
where a small group of animals can be
observed for extensive periods of time
under semi-natural conditions (Atema
eta/,, 1979). Such basic information is
needed to interpret whether, for instance,
a given behavior results from oil
exposure or rather from a combination
of pre-molt aggression and lack of food
Experimental Procedures
Lobsters were trapped locally in the
Woods Hole, Massachusetts, area and
fed herring or mussels while being
laboratory acclimated. Animals which
had molted 2 to 8 weeks prior to
trapping were selected for the experi-
ments to avoid effects of premolt
behavior. Equal numbers of males and
females were distributed between
experimental and control groups. All
lobsters were early adults measuring
from 65 to 75 mm carapace length.
The experiments were conducted in a
continuous flow-through dosing system
with two head tanks (Figure 2). Seawater
inflow to the experimental head tanks
was 4 I/mm. Oil was introduced at a
fixed rate into a fast jet of seawater by a
syringe pump causing rapid emulsifica-
tion. From the head tank, the surface
layer was skimmed off and discarded
and the remaining oil-water mixture
entered six individual 100-liter lobster
tanks. This oil-water mixture is called
the water accommodated fraction
(WAF). The overflow from the individual
tanks ran into a holding tank where
-------
Toxicity Studies
Sublethal interference
with Behavior
Controlled
Laboratory Studies
Semi-controlled
Naturalistic Studies
Field Studies
Chemoreceptor
Physiology
Data Storage
and Processing
Figure 1. Relationship of projects: A multidisciplinary approach provides the context
in which controlled laboratory and toxicity studies may be interpreted.
Control
Experimental
Drain
Figure 2. Flow-through oil dosing system.
other lobsters were exposed to oil for
the neurophysiological studies. The
control head tank, which had an inflow
of 2.6 l/min and no oil, supplied four
100-liter lobster tanks. Inflow to all
individual lobster tanks was kept
between 400 to 460 ml/min. These
tanks had a front glass window for
observation, a shelter and a pebble
substrate. Experiments were carried out
at temperatures ranging from 8° to
23°C in ambient, untreated, and unfil-
tered seawater. The oil, Exxon No. 2 fuel
oil, was obtained from the U.S. Environ-
mental Protection Agency (EPA), Envi-
ronmental Research Laboratory, Nar-
ragansett, Rhode Island. All lobsters
were acclimated to their individual
tanks and experimental conditions for at
least one week prior to starting the
experiments.
Each experiment started with an
additional five days in the tanks serving
as an internal control for the individuals
to be exposed to oil. The experimental
lobsters were exposed to predetermined
concentrations of oil for five days. The
external control group was not exposed
A third 5-day period after the tests
allowed for measurement of possible
recovery or persisting effects. The entire
experiment lasted 15 days.
The behavioral series included nine
tests. Three were performed at oil
concentrations of approximately 0 1
ppm WAF, three at approximately 0.3
ppm, and one at 1.5 ppm. Two additional
tests were done at 0.3 ppm, the first on
lobsters whose antennules were being
cut off during the course of the experi-
ment, and the second, a month later on
the same lobsters, now without the
antennules
The feeding behavior of all lobsters
was recorded twice daily, each day at
the same times in the early morning and
the late afternoon. After one minute of
behavioral observation, food was lowered
to them on a string alternately from the
right and left corners of the tank
Subsequent feeding behavior was
measured by breaking down activity into
five parts. ALERT was the first observable
response to food; WAIT was the period
from then until the lobster left its
shelter; SEARCH was from that point to
its grasping the food, or HIT; and finally
it was noted if the animal actually did
EAT the food. Lobsters were given 10
minutes to complete the sequence
An entirely separate series of neuro-
physiological tests was performed on
lobster antennules to determine the
effects of oil exposure on its chemore-
ceptors. For this experiment, the lateral
flagellum of the antennule was cut off
and placed in a small chamber with
continuously flowing seawater. Test
chemicals were injected into the
seawater flow. Recordings were made
by picking up a small nerve bundle with
a platinum electrode The signal was
amplified, displayed and recorded on
conventional equipment. Experimental
and control antennules were takenfrom
animals during this experiment and also
from animals in the flow-through
holding tank.
The chemical stimuli consisted of five
preparations, mussel juice, No. 2 fuel
oil, mussel juice plus oil, artificial
seawater and, again, mussel juice. The
mussel juice was a mixture of homo-
genized mussel tissue and artificial
seawater, used in concentrations of 1 to
3 ppm. Artificial seawater was the
conventional MBL formula used to
provide consistency. The concentration
-------
of oil was much higher (1 to 3 ppm) than
in the behavioral tests because to
measure chemoreceptor responses one
often needs concentrations about ten
times higher than those needed to see
reactions in the live animal The
sequence of stimuli provided an internal
control for the test of oil effects on
chemoreceptors. The mussel juice
stimulus at the beginning and at the end
showed whether the nerve bundles
were still intact after the test series; the
artificial seawater stimulus applied
midway through the experiment showed
whether they were reacting to mechani-
cal water flow and chemically neutral
stimuli.
Daily samples of water were taken
from the lobster tanks. Infrared (IR) and
ultraviolet (UV) spectroscopic and gas
chromatographic (GC) analyses were
performed to determine the concentra-
tions and types of hydrocarbons present
in both experimental and control lobster
tanks. Salinity, pH, oxygen content and
ammonia were monitored every other
day during the first experiment.
Results and Discussion
Results from the chemical analyses of
lobster tank water show that experi-
mental tanks received petroleum at the
approximate rates intended, and that
the oil cleared from the water quickly
after inflow stopped. Neither control nor
experimental tanks carried significant
amounts of petroleum before the tests
(Figure 3). The IR-observed background
of about 0.05 ppm consisted mostly of
non-petroleum lipids. Behavioral results
did not show observable differences
between experiments with fluctuating
as compared with constant petroleum
dosing. Other water quality criteria (pH,
oxygen, ammonia, and salinity) remained
at satisfactory levels. Thus, the added oil
is shown to have been the primary
variable in these experiments
The behavioral results fall into three
groups which together bracket the
sublethal levels of exposure. The 0.1
ppm level caused little observable
effects during the oil exposure (days 6 to
10). The 1.5 ppm level resulted in gross
neuromuscular defects which appeared
after a few hours and lasted for several
days after exposure had stopped. Five
out of six such animals could hardly
walk and twitched in cramped postures;
they responded poorly or inappropriately
to food. This level approaches lethal
effects; in nature such animals would
be helpless and could be described as
"ecologically dead." This leaves the
relatively narrow range of 0.1 to 1.0
ppm as the oil concentrations which
cause sublethal effects. All experiments
at the exposure level of 0.3 ppm showed
lobsters less likely to look for food and
eat after 6 to 24 hours of exposure to oil.
They usually recovered after one day in
clean water Behavioral changes were
apparent in both the occurrence and
duration of feeding behaviors. Some
Expt. 2
Exptl. Tank, Pre-oil
Amp. x Scale, 10 x 2
Expt. 2
Control Tank, Pre-oil
Amp, x Scale. 10 x 2
Expt. 2
Exptl. Tank, Oiled
Amp. x Scale, 10x8
Expt. 6
Control Tank
Amp. x Scale, 1O x 2
Expt. 6
Exptl. Tank, Oiled
Amp. x Scale, JO x 2
225 200
150
100
Figure 3.
Gas chromatograms indicating lack of petroleum lipids in control tanks and
in experimental tanks before oil was added, and the presence of typical
Number 2 fuel oil peaks in experimental tanks during oil exposure at
0.1 ppm and 0.3 ppm.
-------
animals failed to ALERT to food; others
SEARCHED, but did not find the food.
The result was that most lobsters
missed several feeding opportunities
during the exposure period. Similar
effects were observed at high and low
temperatures. Lower temperature and
lack of antennules did not change the
effect of 0.3 ppm oil exposure on lobster
feeding behavior.
The neurophysiological experiments
on antennular chemoreceptors show
that 1 ppm WAF No. 2 fuel oil is a
chemical stimulus and interferes with
the normal response to a common food
odor, mussel juice (Figure 4). Because
antennules of control and experimental
lobsters react alike, chemoreceptor
interference is probably not the only way
in which petroleum exposure disrupts
lobster behavior. Previous experiments
(Atema, 1976) suggest that oil can serve
as a chemical attractant, repellent and
general neurotoxin, meaning that
lobsters may be subjected to a mixture
of chemical stimuli in the presence of
food and petroleum. This may well
cause sensory "confusion" and hence
inappropriate responses. These could
combine with general neurotoxic effects,
which may render the animal unmoti-
vated, uncoordinated and unresponsive.
Data from the behavioral studies
support this hypothesis. If antennular
chemoreceptors were the only mediators
controlling feeding behavior, one would
expect animals without antennules to
react less to oil exposure than those
with antennules. One experiment
indicates otherwise, suggesting either
that chemoreceptors from another part
of the lobster, e.g., legs, take over the
function of the antennular chemore-
ceptors, or that oil exposure affects the
lobster's motivational state.
The inconsistent behavior of lobsters
exposed to a critical concentration of oil
argues against purely chemosensory
interference, but for an effect on
motivational state On different days,
exposed lobsters reacted differently to
the presence of food. Those animals
that one day alerted normally to the
presence of food (an activity more likely
under chemosensory control) also
searched and found the food that day
(activities more likely under motivational
control), whereas those that did not
show ALERT behavior also did not
search and eat. The physiological basis
for the observed lack of motivation is not
known, but may be caused by general
neurotoxicity of certain concentrations
of petroleum These effects appear
Oil
(a)
Mussel
(d) |
7 sec
Figure 4. Neurophysiological responses of antennular chemoreceptors to chemical
stimuli.
quickly and obviously at higher exposure
levels (>1 ppm of oil) when lobsters
lose neuromuscular control.
There apparently is a limited range
(between 0.1 and 1 ppm) in which oil
toxicity is sublethal for the American
lobster Clear effects on feeding behavior
are measured consistently at 0.3 ppm
The ecological implication of impaired
feeding behavior is a loss of survival
fitness in competition with other
animals.
Conclusions and
Recommendations
1) Five days of exposure to 0.3 ppm
WAF No. 2 fuel oil caused consis-
tent sublethal interference with
feeding behavior of adult and
subadult lobsters.
2) Feeding interference lasted for
one day after 5-day exposure to
0.3 ppm.
3) Sublethal interference with lobster
behavior was not observed at 0.1
ppm exposures, however, at 1 to 2
ppm exposures severe neuromus-
cular abnormalities appeared,
leading within a few hours to
cramped postures, spastic behavior
and unresponsiveness, and even-
tually death.
4) Thirty hours of exposure to 1.5
ppm WAF No. 2 fuel oil caused
cessation of feeding for over 6
days in most lobsters.
5) Lobsters smell the presence of 3
ppm fuel oil; furthermore, some of
their chemoreceptors showed
modified responses to food odors
in the presence of 3 ppm fuel oil,
indicating that oil interferes with
their smelling food.
6) Feeding behavior interference
may be caused partly by oil effects
on chemoreception and by oil-
induced changes in feeding moti-
vation, i.e., hunger and/or fear.
7) Based on these results and a
safety factor of 10, short-term
exposure to water column levels
of 0.01 ppm WAF No. 2 fuel oil may
not interfere with feeding behavior
of adult lobsters. However, further
investigations involving larval
stages and long-term exposures
are recommended before safety
levels for chronic petroleum pollu-
tion are firmly established. Specific
behaviors such as hatching, mol-
ting, growth, feeding, phototaxis,
settling and substrate selection
should be studied
References
Atema, J. Sublethal effects of petroleum
fractions on the behavior of the
lobster, Homarus americanus, and
the mud snail, Nassarius obsoletus.
In: M. Wiley (ed.), Estuarine Pro-
cesses, Vol I. Uses, Stresses, and
Adaptation to the Estuary. Academic
Press, Inc. New York, NY, pp 302-
312, 1976.
Atema, J , S. Jacobson, E. Karnofsky, S.
Oleszko-Szuts, and L Stein. Pair
formation in the lobster, Homarus
americanus: behavioral development,
pheromones, and mating. Mar. Behav.
Physio/., 6:277-296, 1979.
-------
Atema, J. and L Stein. Effects of crude
oil on the feeding behavior of the
lobster, Homarus americanus. Environ.
Pollut, 6:77-86, 1974.
Blumer, M. Oil contamination and the
living resources of the sea. FAO
Technical Conference on Marine
Pollution and Its Effects on Living
Resources and Fishing. Rome, Italy,
December 9-18, 1970.
Sanders, H.L, J.F. Grassle, G.R Hamp-
son, LS. Morse, S. Garner-Price, and
C.C. Jones. Anatomy of an oil spill:
long-term effects from the grounding
of the barge Florida off West Fa Imouth,
Massachusetts. J. Mar. Res., 35.265-
280, 1980.
Jelle Atema, E. B. Karnofsky. S. Olszko-Szuts, and B. Bryant are with Boston
University, Marine Biological Laboratory, Woods Hole, MA 02543.
Don C. Miller is the EPA Project Officer (see below).
The complete report, entitled "Sublethal Effects of Number 2 Fuel Oil on Lobster
Behavior and Chemoreception," (Order No. PB 82-192 444; Cost: $9.00,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 221'61
Telephone; 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Research Laboratory
U.S. Environmental Protection Agency
South Ferry Road
Narragansett, Rl 02882
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
US ENVIK PKOTtCTlUN
KttiON 5 LlbRAKY
230 S DEARBORN STRtEF
CHICAGO 1L 60604
------- |