MFR PAPER 1278
Flavors in Fish From Petroleum Pickup
MAURICE E. STANSBY
ABSTRACT - All flavors noted in fish resembling petroleum oil are not derivedfrom oil in water. Origins of various flavors found in fish are discussed. Many factsneeded to understand the relationships between levels ofpetroleum in water and infish and flavor of the fish have not been elucidated. Particularly, there are largediscrepancies between reports by different investigators regarding levels of hydrocarbons and resulting flavor intensities. More research is needed to resolvethese contradictions.
least to the effect that there might beflavor problems.
Before discussion of research whichhas been carried out on matters specificto our interest, a short introductory section is included to give background ondifferent types of flavors occurring infish and from what sources they arederived. This information is needed inorder to understand the differentiationbetween baseline flavor of fish in theabsence of petroleum from flavors directly connected with the presence ofpetroleum.
GENERAL NATUREOF FLAVORS
AND ODORS IN FISH
INTRODUCTION
It has been common knowledge formany years that when fish are caughtand stored aboard the fishing vessel inice, and bilge water containing oil accidentally comes in contact with the fish,the quality of the fish is impaired from aresulting off-flavor development in theflesh. The off-flavor can range fromonly a slight impairment, in caseswhere only small amounts of oil areinvolved, to a condition where the fishare completely ruined for human consumption. This kind of accidental contamination of fish has occurred frequently enough that there can be nodoubt that when fish are stored withaccess to sufficient petroleum oil, adamaging alteration in flavor, whichhas generally been designated as a taint,develops.
In such cases as have just been described, the fish are dead at the timethey have been in contact with watercontaining petroleum. We cannot justifiably conclude from such incidentsthat live fish living within waters contaminated by petroleum would developsuch a taint. It appears quite likely thatin the case of live fish, at least a part, ifnot all, of any petroleum entering thefish would, by the processes ofmetabolism, be altered, perhaps excreted so that there might be little or noalteration in the fish's flavor. It is thepurpose of this paper to consider research findings dealing with the pick-upof petroleum by fish living in
January 1978
petroleum-contaminated waters withreference to alteration in flavor of theirflesh.
Only a relatively small amount ofresearch has been carried out on suchmatters as the effect of petroleum uponf1a vor of fish I iv ing ina petroleum contaminated environment. This ispartly because interest in such environmental problems has not, untilquite recently, been evident. A furtherhindrance to such research is caused bythe rather difficult nature of suchstudies. Flavor of fish, or of any foodmaterial, is a completely subjecti vemeasurement which does not readilylend itself to use of scientifictechniques. For direct measurement offlavors, it is necessary to use highlytrained flavor panels and even when thisis done, the subjective nature of thetests often leaves questions as to theaccuracy of the results. On the otherhand, it is possible to study flavors in anindirect manner by chemically measuring the components responsible for thedifferent flavors. In the present instance, for example, a comparison ofthe content of hydrocarbons known tooccur in petroleum found in fish livingin petroleum-contaminated waterwould give presumptive evidence, at
Maurice E. Stansby is with theNorthwest and Alaska FisheriesCenter, National Marine FisheriesSerVice, NOAA, 2725 MontlakeBlvd. East, Seattle, WA 98112.
Three types of flavors, based uponthe source from which they are derived,occur in fish, and each of these will bedescribed briefly. These are naturalflavors, unnatural flavors, and decomposition flavors.
Each species of fish contains manydifferent flavor chemical componentswhich blend together to give thatspecies a particular flavor. Just as suchother flesh foods as beef, pork, lamb,and chicken have composite flavorswhich enable us to distinguish one fromanother, so also such species as salmon,herring, and cod have different naturalflavors, each characteristic for the different species. The flavoring components characteristic of each speciesoccur both in the oil and in the proteinof the fish. Ordinarily the most flavorfulcomponents are chemical entitieswhich are dissolved in the oil of thefish. Nevertheless, fish protein alsopossesses some characteristic, thoughoften quite bland, flavors.
Unnatural flavors occur in occasionalspecimens of a species which give thesefish flavors not typical for the species.Such flavors occur especially when fishare consuming some type of feed whichthey normally do not eat. This may giverise to abnormal flavors, which sometimes are quite pronounced and whichmay be undesirable. For example, during certain seasons flatfish, such assole, feed upon certain plankton whichgive rise to a flavor reminiscent of thatof iodine. In other cases fish may consume, during a small portion of their
13
life span, feed giving a wide variety ofabnormal flavors. In some cases, suchflavors might resemble that of petroleum, even though no petroleum was involved. In other cases, the fish mayacquire unnatural flavors by coming incontact with substances accidentally introduced into the water, petroleumbeing an example of this type of source.The fish could acquire such material byingesting it through the digestive tract,by taking it into their respiratory systemthrough the gills, or perhaps in somecases by direct absorption through theskin.
Decomposition of fish is the thirdprocess which can lead to off odors andflavors in fish. Bacterial decomposition(spoilage) leads to production of mostlyvolatile components like trimethylamine, ammonia, and hydrogensulfide. These result in off odors, butduring cooking of the fish most of thecompounds responsible for the odorsare volatilized so that the flavor of thecooked fish is altered much less thanmight have been indicated from odorobservations. The components responsible for the natural flavors characteristic of the various species are highlylabile and such flavors disappearrapidly, due probably to oxidation, asthe fish are held at refrigerated temperatures.
Lipids (mostly oils) of fish areoxidized to some extent during storagein ice and to an extensi ve degree for fishfrozen and held in cold storage. Theoxidation products, which are mostlycarbonyls, give rise to rancid types offlavors.
For research on flavors in fish themost common approach used by scientists has been to attempt to isolate fromthe fish chemical constituents mostprobably responsible for the flavor.Such research is time consuming and itinvolves intricate separation procedures to isolate the chemical constituents making up the flavor fromhundreds or thousands of other compounds which are also present. Theflavor components occur ordinarily inonly trace amounts, which makes itmost difficult to separate them from theother components, many of which arepresent in amounts several orders ofmagnitude greater than the flavor components. Identification of the isolated
14
components is usually confirmed bymass spectrometry, sometimes used inconjunction with other criteria.
Many investigations do not gobeyond the stage of isolation and identification of components which theorywould indicate are likely causes of theflavor. In order to give the researchresults any real meaning, it is necessaryto confirm that the chemical components isolated and identified really dopossess the flavor of the fish which isnoticed when the cooked fish is eaten.In order to carry out this additional research it is necessary to use a highlytrained sensory evaluation panel of experts who are thoroughly familiar withthe different flavors associated withfish. Such a panel can positivelyconfirm whether the flavor of the fish iscaused by the mixture of isolated andidentified components and also can establish the ranges of concentration ofthe different components which makeup the flavor. Unless this final step of aflavor evaluation is carried out, the results of a purely chemical study may beof little value from any practical application standpoint.
RESEARCH ONFLAVORS PICKED
UP FROM PETROLEUM
The number of investigations andpapers in this field is not huge and sincemost of the articles that have appeareddeal with some limited aspect such asanalytical methods, or are of a verygeneral nature, the number of researchprojects set up to learn the nature of thephenomena involved or looking at theoverall picture are very few indeed,Most of this review will deal with theselatter types of research projects.
Some of the most comprehensive research in this field is coming from Torry Research Station (Mackie et aI.,1972; Hardy et aI., 1974; Howgate etaI., 1976). These investigations include, for example, research on hydrocarbons isolated from brown troutcaught in a stream which had been polluted by a 2,000 gallon diesel oil spill.The hydrocarbons were separatedchemically from other constituents ofthe fish and were shown to be identicalto hydrocarbons contained in the dieseloil. Furthermore, the odors and flavorsof the hydrocarbons isolated from the
fish matched exactly those from thehydrocarbons from samples from thesame batches of diesel oil as had beenspilled in the stream. Hydrocarbons insmall trace amounts were isolated frombrown trout caught in the same streamat sufficient distance upstream from thesite of the spill so that no petroleum waspresent in the water. The odor of thesehydrocarbons had no similarity to thosefrom the polluted fish and weredoubtlessly of biogenic origin. Theodor and flavor of the hydrocarbonsfrom these control fish were almost niland were not at all similar to that fromthe hydrocarbons from the oil pollutedfish. In another set of experiments byTorry Research Laboratory investigators (Hardy et aI., J974) the rate ofuptake of hydrocarbons into the liver ofwild codlings was measured. The fishwere kept in tanks of seawater at thelaboratory and the fish were fed codliver oil containing Kuwait crude petroleum. As controls in another batch offish, the oil fed was entirely cod liveroil. In these experiments there was apreferential absorption of carbon chainlength of around 26 from the fish fed thepetroleum but not from the controls.Hydrocarbons of chain length CIS toCZI were absorbed to an insignificantextent. In the hydrocarbon chain lengthrange from C 22 to C26 there was anincreasing amount absorbed up to amaximum at chain length C26 . Fromchain length C26 to C30 there was adecreasing extent of absorption down tochain length C31 , at which no significant absorption into the codling livers took place. There was in no instanceany evidence of preferential absorptionby odd as contrasted to even hydrocarbon chain lengths. The first preliminaryexperiment, while too limited in scopeto be conclusive, indicates that the codling possesses ability to preferentiallyabsorb hydrocarbons of certain chainlength. During a 6-month period aftercessation of feeding of the crude petroleum by the fish, hydrocarbon levelshad fallen to about half the maximumlevels reached but were still about 65times higher than in'the control samples.
In another investigation, scientists inAustralia's CSIRO (CommonwealthScientific and Industrial Research Organization) Division of Food Preserva-
Marine Fisheries Review
tion have been approaching this problem in a somewhat different way. Forexample, Shipton et al. (1970) investigated a kerosene-like flavor in mullet.Upon solvent extraction of such fishand examination and separation of hydrocarbons, they found characteristicssimilar to those of hydrocarbons foundin crude oil. In other cases, however,Vale et al. (1970) found compoundsapparently responsible for thekerosene-like flavors which arose fromthermal decomposition of naturally occurring components in fish.
Deshimaru (1971) studied the effecton fish flavor of rearing yellowtail,Seriola dorsalis, in water containingcrude petroleum oil. The amount of thecrude oil in the fish rearing tanks wasset attwo levels, 10 ppm and 50 ppm, indifferent portions of the experiment. Athird group of fish was held in watercontaining no petroleum but was fedfeed consisting of fish containing I percent crude petroleum oil. The meatfrom the fish reared with 50 ppm oil inthe water had a strong fishy flavor if thefish had been exposed to the oilcontaining water for at least 5 days. Nofish odor was observed, however, evenafter 13 days rearing with the oil whenthe level of oil in the water was onl y 10ppm. A fishy flavor developed aftermore than 13 days exposure, but only toa slight degree. When the fish were fedwith oil at the I percent level in thefeed, a fishy flavor developed at a verylow level after 5 days under test, but thedegree of intensity had not increased bythe end of the experiment (13 days).The presence of a CZ7 paraffin hydrocarbon was identified by gaschromatograph analysis of the flesh ofyellowtail for the samples that had beenheld in water containing both 50 ppmand 10 ppm levels of petroleum.
Wilder l carried out quite differenttypes of experiments on lobster meatand lobster liver (tomalley). In theseexperiments the living lobsters werecoated with a fairly heavy layer ofBunker C petroleum oil. The lobsterswere then, in some cases, returned toseawater in a 48-gallon tank in which
'Wilder, D. G. 1970. The tainting of lobster meatby Bunker C oil alone or in combination with thedispersant, Corexil. Unpubl. manuscr., 23 p.Fish. Res. Board Can. BioI. Stn., Sl. Andrews,N.B.
January /978
the water was replaced at the rate of 1.6gallons per minute. After exposure torunning seawater in this way, lobsterswere observed for disappearance of theoil. Within 6 hours less than 10 percentof the oil remained on the lobsters.After 96 hours only a few droplets of oilwere visible at crevices on the ventralsurface. Lobsters were then cooked andboth the flesh and the liver examined bya taste panel. After 192 hours, morelobsters were withdrawn and subjectedto the same treatment. In neither themeat nor the liver, after either 4 or 8days exposure to the running seawater,was there any evidence of taint foundby the panel. In another part of theexperiment lobsters were fed bait covered with Bunker C oil. The lobstersreadily consumed the bait, but againthere was no evidence of taint after 4 or8 days exposure to running seawater.
On the other hand, when untreatedlobster were held in a tank of seawatercontaining I part of Bunker C oil perI ,000 parts of water ( 1,000 ppm) for 90hours, both the meat and liver of thelobsters after cooking had a very decidely objectionable oily flavor. Theliver samples had a stronger flavor thandid the lobster meat. When the lobstertreated in this way were held live inrunning seawater for as long as 3weeks, the oily taste in the meat andliver, after the lobsters were cooked,still remained although at reduced intensity.
Some additional tests indicated thatoil could be more completely removedfrom lobster by wiping if the dispersant, Corexit,2 is used in wateremployed for the deoiling process.
A classical example of a petroleumlike odor in fish, which upon investigation proved to be something quite unrelated to petroleum, was documented byMotohiro (1962). As much as 5 percentof the canned chum salmon producedby the Japanese in the Bering Sea possessed a distinctly petroleum-like odor.After an extensive investigation, it wasfound that the odor was from dimethylsulfide which came from the precursor,dimethyl-2-carboxyethyl sulfoniumsulfide. This substance was found in the
'Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA.
digestive tract of a portion of the chumsalmon packed in the north Pacific andit is believed to arise from certain algaeconsumed on occasion by salmon. Although the odor of the fish had whatmost observers described as a petroleum-type odor, it was shown conclusively that no petroleum product wasinvolved in formation of the odor.
Mullet, possessing a petroleum typeof flavor, have in some cases beenfound among the Australian catch. In apreliminary investigation, Grant (1969)suggested that, like the petroleumflavor found in canned chum salmon,this flavor in mullet might stem fromdecomposition of feed components toproduce dimethylsulfide. In more extensive investigations, Connell (1971,1974) showed that this mullet off-flavorarose from petroleum contamination.The volatile constituents from thetainted mullet were steam distilled andexamined by gas chromatography-massspectrometry. The resulting gaschromatograms and mass spectrashowed close resemblance to thosefrom commercial kerosene. It was alsofound that sediments in the BrisbaneRiver in which the mullet spent part oftheir time also contained substanceshaving gas chromatography-mass spectrometry characteristics similar tokerosene. It was concluded that themullet probably included some of thesesediments with their feed although theymay have picked it up from the water.
Nitta et al. (1965) investigated thedisagreeable odors found in fish occurring near petroleum refining plants atYokkaichi, Japan. It was found that theodor of the fish flesh was similar to theodor of the effluent from the petroleumrefineries. The investigation revealedthat the odor could be picked up eitherfrom the water in the effluent or frombottom muds near the refinery. Theconcentrations of ether extract substances which were able to produce thiseffect were 0.01 ppm in water or 0.2percent in the muds. It was determinedthat the substances causing the odorsentered the fish through the respiratorysystem (gills). The odor became apparent in the fish flesh after only about 24hours exposure to the effluents. Fish didnot avoid the oily effluents; on the contrary, the fish seemed to be attracted bythem.
/5
Other investigations correlatingpetroleum-derived flavor or odor of fishwith petroleum refining operationshave been carried out (Krishnaswamiand Kupchanko, 1969; Miyake, 1967;Ogata and Miyake, 1970; Yoshida andUezumi, 1961; Yoshida et aI., 1967).Of particular interest here, however, isthe relationship between offshore drilling and oil taint (Mackin and Sparks,1962; St. Amant, 1958; Menzee· 4).The State of Louisiana (St. Amant,1958) tabulated 60 complaints of oystermen with operations near offshoreoil wells with about 40 percent of thecomplaints dealing with oily taint. Investigations showed that mud at oysterbeds near such drilling operations hadvery high ether soluble extracts (up to14,000 ppm). Usually the content ofsuch extractives decreased as onemoved away from the vicinity of thedrilling operation, but in a few instances maximum oil contamination ofthe mud was at a distance of 500 feet ormore from the site of the wells. It wasfound that this situation usually occurred with wells deeper than 10,000 feetand especially where the drilling operation employed a mixture of diesel oilwith drilling muds. It was found thatsuch drilling operations could result intainted oysters if the oyster beds werewithin 5,000 feet of the wells in closedchannels or within 2,500 feet in openbays. The oily taste in such oysters wasfound to persist for as long as 4-6months after oil pollution ceased.
Oily taste from petroleum has beenreported for shellfish other than oysters.Such reports have dealt with mussels(Nelson-Smith, 1970; Brunies, 1971)and clams (Hawkes, 1961).
GENERAL DISCUSSION
The amount of research carried outon the development of an off flavor,sometimes designated as taint, by absorption of petroleum oil or its components is not at all extensive. Fairlynumerous articles have dealt, often
'Menzel, R. W. 1947. Observations and conclusions on the oily lasting oysters near the tankbanery of the Texas Co. in Crooked Bayou at BaySte. Elaine. Unpubl. rep., 6 p. Tex. A&M Res.Found., Proj. 9.'Menzel, R. W. 1948. Report on two cases of"oily tasting" oysters at Bay Ste. Elaine oilfield.Unpubl. rep., 9 p. Tex. A&M Res. Found., Proj.9.
16
superficially, with the presence of suchflavors in fish, but very few researchpapers have been published in whichthe subject is investigated, even covering only a small aspect, in any kind ofthorough and systematic way. I couldfind no research results on developmentof such flavors under Arctic conditions.
Petroleum-like flavors have beenwidely reported to occur in fish, butfrequently when such flavors are furtherinvestigated they are found to be derived not from any petroleum precursorbut rather from some alteration innaturally occurring components of thefish. Nevertheless enough instances ofpetroleum-like flavors in fish have beeninvestigated and fully documented thatthe flavor arose from a petroleum product or one of its components, to leaveno doubt that such products definitelycan give rise to off flavors in fish.
Taint type flavors can occur by merecontact of the surface of the fish withpetroleum. This occurs only when relatively large quantities of petroleum areinvolved. The petroleum and its flavorcan in such cases be completely removed, or nearly so, by washing inrunning water with the removal facilitated by use of a detergent. In the moreusual case the petroleum is picked upfrom the water by the fish either directlyfrom the water through the respiratorysystem or from feed through the digestive tract. In such cases much less oil isrequired to bring about the effect andthe off flavor is removed only by removing the fish from a source of oil.The pickup of oil in these cases is usually a slow process requiring days orweeks of exposure and reduction andelimination of the flavor is an evenslower process requiring weeks ormonths of residence of the fish in anoil-free environment for complete removal of flavor or odor.
The vast predominance of research inwhich fish are exposed to waters containing petroleum or its componentsand then, after suitable intervals, areexamined for flavor pickup have usedrelatively high concentrations of oil inthe water, far more than ordinarilywould occur under normal circumstances in the natural environmentexcept under most unusual circumstances. Levels used generally areof the order of parts per million. For
example Deshimaru (1971) exposedfish to water containing either 10 ppmor 50 ppm oil and fed fish feed containing 10,000 ppm of oil; Wilder (seefootnote)) exposed lobsters in tanks ofwater containing I ,000 ppm oil. Thelevel of oil in waters occurring as aresult of discharge of oil by industrialoperations are many orders of magnitude lower than such levels. Ordinarily, levels are at parts per billion, evenat parts per trillion, concentration. Theonly cases where many parts per million levels would occur are immediately adjacent to a massive oil spillor perhaps at a discharge pipe comingdirectly from a processing or manufacturing plant. Even then such levels ofthe order of parts per million are verytransitory and owing to dilution soondiminish to lower levels. Because ofthese circumstances the results of mostof the research reported in the literatureon taint-like flavor picked up in controlled experiments does not necessarilyreflect what happens in the usual environment. Based upon the degree andrate of development of taint in whatcontrolled experiments as have beenreported and where far greater levels ofoil were used than would be encountered in open natural waters, it is thebelief of this reviewer that taint frompetroleum oil would never result fromfish living in open ocean waters (barring some major catastrophe such as theTorrey Canyon incident, and then onlyfor a short time span). Taint is a common occurence when fish are accidentally dipped in oil or oil-water emulsions such as might occur in a restrictedhold aboard a fishing vessel. It probablyalso occurs occasionally in streams onvery restricted bays or harbors wherelarge volumes of oil accidentally comein contact with the fish. There is noevidence for its occurrence either forfish in the open sea or even in large bayswith moderate to considerable discharge of oil from industrial operations.
Information on the level ofpetroleum-derived components in fishwhich barely are sufficiently high tocause detectable flavor are still too incomplete to give any indication of whatlevels might be tolerable. Ineson andPackham (1967) report threshold odorconcentrations for certain petroleumproducts in water. These range around
Marine Fisheries Review
MFR Paper 1278. From Marine Fisheries Review, Vol. 40, No.1, January 1978.Copies of this paper, in limited numbers, are available from 0822, User Services Branch, Environmental Science Information Center, NOAA, Rockville,MD 20852. Copies of Marine Fisheries Review are available from the Superintendent of Documents, U.S. Government Printing Office, Washington, DC20402 for $1.10 each.
100 ppb to 2 ppm for most crude orrefined petroleum oil. Research isneeded to yield information onmaximum levels of such compoundsthat can occur in fish without resultingin any perceivable off odor or flavor.
What, then, do we know, based uponthe few experiments described? Withthe limitations that most of the experimental work has dealt with concentration of petroleum several orders ofmagnitude greater than occurs in openwaters and that none of the research wascarried out under Arctic conditions, thefollowing conclusions can at least tentatively be reached.
Petroleum reaches the marine organisms primarily through the watercolumn, but may come from heavilycontaminated mud or other sedimentarybottom. A major mode of entry isthrough the respiratory system, although it can also be picked up in thefeed and there is no evidence at all torule out that the food chain may not beof equal importance as a mode of entry.Taint types of flavors have been reported in at least one instance (Nitta etaI., 1965) with as little as 0.01 ppmpetroleum in water which has been incontact with fish for no longer than 24hours.
Fish preferentially absorb hydrocarbons with chain length in the range ofC22 to C30 with a maximum at C26 , andodd and even carbon chain length hydrocarbons are absorbed at about thesame rates. There is no agreementamong investigators as to the levels ofpetroleum in water needed to bringabout off flavors in the meat of the fish.Thus Deshimaru (1971) reported thatfish in waters containing 10 ppm ofpetroleum within 13 days exposure developed no fishy odor, yet Nitta et al.(1965), in experiments in natural waters, reported definite off odors developed in fish exposed for only 24hours to levels of petroleum of only0.0 I ppm. While differences in type ofpetroleum and varieties of fish in different instances doubtlessly are important, such wide differences in thresholdconcentrations of three orders of magnitude plus 13 times difference in exposure time seem hard to account for onany such basis.
Retention of off odors after removalfrom the pollution may require as long
as 6 months or more before disappearance of the off flavor. On the otherhand, if the contamination occurs onlyas a result of external application, as inthe experiments of Wilder (see footnoteI) the off-flavored meat of the fish canbe returned to normal after only a fewhours holding in pure running water.
Although several investigators haveisolated, from fish or other marine organisms which possessed an off or taintodor, compounds having chemicalcharacteristics closely resembling petroleum hydrocarbons, there has beenno identification of single compoundsnor groups of mixed, identified compounds which possess the characteristic"taint" flavor resulting from exposureof marine organisms to petroleum.
The area of research in this fieldwhich has been most neglected and inwhich more work should certainly beundertaken is a definitive evaluation ofthe maximum concentration of different kinds of petroleum in water inwhich fish can reside without pickingup petroleum types of flavors. In whatfew reports on this matter exist in theliterature there is such a complete disagreement among .levels found whichare said to lead to petroleum flavors orodors in the fish that we really knownothing definite.
A recent research trend has been toward looking at what happens to fishwhen they live at the bottom near sediments which contain oil. At Torry Research Laboratory (Howgate et aI.,1976), research is at an early stage onlooking at effects on flavor of fishwhich have been kept close to mixturesof sand and petroleum under laboratoryholding conditions. Work on more general effects of petroleum on fish heldunder similar laboratory conditions isunderway at the Northwest and AlaskaFisheries Center. Most research upuntil recently has involved effect of oilsuspended or dissolved in the watercolumn. It will be interesting to findwhat differences occur between effectof oil in sediments and oil in the watercolumn as such research proceeds.
LITERATURE CITED
Brunies, A. 1971. Taint of mineral oil in mussels. Arch.Lebensmitlelhyg., March 1971, p. 63-64.
Connell, D. W. 1971. Kerosene-like tainting in Australian mullet. Mar. PoilU!. Bull. 2: 188-190.
____. 1974. A kerosene-like tainl in the seamullet, Mugil cephalus (Linnaeus). I. Compositionand enviromnental occurrence of the tainting substance. Aust. J. Mar. Freshwater Res. 25:7-24.
Deshimaru, O. 1971. Studies on the pollution of fishmeat by mineral oilr-I. Deposition of crude oil infish meat and its detection. Bull. Jap. Soc. Sci. Fish.37(4):297-301. (Fish. Res. Board Can. Transl. Ser.2038-1972.)
Grant, E. M. 1969. "Kerosene" taint in sea mullet(Mugil cephalus Linnaeus). Fish. Notes 3: 1-13.
Hardy, R., P. R. Mackie, K. J. Whittle, and A. D.McIntyre. 1974. Discrimination in the assimilation ofn-aI.kanes in fish. Nature (Lond.) 252:577-578.
Hawkes, A. L. 1961. A review of the nature and extentof damage caused by oil pollution at sea. Trans. N.Am. WildJ. Nat. Resour. Conf. 26:343-355.
Howgate, P., P. R. Mackie, K. J. Whittle, J. Farmer, A.D. McIntyre, and A. Eleftheriou. 1976. Petroleumtainting in fish. Proc. Int. Counc. Explor. Sea Workshop, Petroleum Hydrocarbons in the Environment,Aberdeen, Scotland, Sept. 9-15, 1975.
Ineson, J., and R. F. Packham. 1967. Contamination ofwater by petroleum products. In P. Hepple (e1itor),The joint problems of the oil and water industries, p.97-116. Inst. Petrol., Lond.
Krishnaswami, S. K., and E. E. Kupchanko. 1969.Relationship between odor of petroleum refinerywastewater and occurrence of "oily" taste-llavor inrainbow trout, Salrno gairdneri. J. Water Pollut. Control Fed. 41:RI89-RI96.
Mackie, P. R., A. S. McGill, and R. Hardy. 1972.Diesel oil contamination of brown trout (Sa/rno trunaL.). Environ. Pollut. 3:9-16.
Mackin, J. G., and A. K. Sparks. 1962. A study of theeffect on oysters of crude oil loss from a wild well.PubI. Inst. Mar. Sci., Univ. Tex_ 7:230-261.
Miyake, Y. 1967. Studies on abnormal odor fish by oilpollution in Mizushima Industrial Area I. Distributionof abnormal odor fish. Okayama Igakkai Zasshi81(3-4): 193-197.
Motohiro, T. 1962. Studies On the petroleum odor incanned chum salmon. Mem. Fac. Fish. HokkaidoUniv. 10:1-65.
Nelson-Smith, A. 1970. The problem of oil pollution ofthe sea. Adv. Mar. BioI. 8:215-306.
Nitta, T., K. Arakawa, K. Okubo, T. Okubo, and K.Tabata. 1965. Studies on the problems of offensiveodor in fish caused by wastes from petroleum industries. Bull. Tokai Reg. Fish. Res. Lab. 42:23-37.
Ogata. M., and Y. Miyake. 1970. Offensive odor substance in fish in the sea along petrochemical industries. Jap. J. Public Health 17:1125-1130.
51. Amant, L. S. 1958. Investigation of oily taste inoysters caused by oil drilling operations. SeventhBienn. Rep., La. Wildl. Fish Comm., p. 75-77.
Shipton, J., J. H. Last, K. E. Murray, and G. L. Vale.1970. Studies on a kerosene-like taint in mullet(Mugi/ cephalus). n.-Chemical nature of the volatile constituents. J. Sci. Food Agric. 21:433-436.
Vale, G. L., G. S. Sidhu, W. A. Montgomery, and A.R. Johnson. 1970. Studies on a kerosene-like taint inmullet (Mugi/ cephalus). I.-General nature of taint.J. Sci. Food Agric. 21:429-432.
Yoshida, K., and N. Uezumi. 1961. On the problem ofoffensive-odor fish in petrochemical Industrial AreaI. Hyg. Life (Japan) 5(4):11-17.
____, • and K. Kosarna. 1967. Onthe problem of offensive-odor fish in petrochemicalIndustrial Area II. Hyg. Life (Japan) 11(2):8-13
January 1978 17