Sign In Sign-Up

Welcome!

Close

Would you like to make this site your homepage? It's fast and easy...

Yes, Please make this my home page!

No Thanks

  Don't show this to me again.

Close

This is my lit review as of June 1, 2002:

A review of taste perception emphasizing metallic attribute perception
by John Smythe

     To understand perception of oral stimuli, one should consider the physiology of the sensory organs in the tongue. Plattig (1988) reports that the taste pore is an opening in the papillae approximately 2 µm in diameter. From the taste pore, microvilli of 0.1-0.2 µm diameter and 1-2 µm length extend into the oral cavity. After the diffusion of taste-active compounds through the saliva, these microvilli are believed to bear chemical receptors that bind taste molecules in the initial stages of taste perception. These receptor cells bind taste molecules and start the cascading sequence of events that leads to perception. Basal cells exist, as well, with the apparent purpose of regenerating taste cells.
     The terminal branchings of myelinated nerve fibers are known as telodendria. The diameter of these fibers ranges from 1-6 µm. These, combined with unmyelinated fibers of 0.3-1 µm in diameter, enervate the taste buds. The unmyelinated fibers are thought to carry information of the vegetative nerve system to the taste cells in an efferent system. In contrast, the telondria are believed to transmit the gustatory afferent information to the central nervous system. These signals are carried to the brain via the seventh and the ninth cranial nerves. The seventh cranial nerve, N. VII, is known as the facialis nerve and is responsible for stimuli on the anterior two thirds of the tongue. The ninth cranial nerve, N. IX, is known as the glossopharyngeus nerve, and is responsible for signals on the posterior one-third of the tongue. There also exist pharyngeal and oesophageal taste buds, but these are generally connected to the tenth cranial nerve, N. X, known as the vagus nerve (Plattig, 1988). Unfortunately, as of yet there exists no reported experimental evidence investigating the taste response of individual human nerve fibers in relation to either the peripheral or central nervous system (Delwiche, 1996).
     Boudreau (1986) offers suggestions based on animal studies regarding what type of taste information is carried by the various cranial nerves. He suggests that the trigeminal nerve, N. V, carries astringent, pungent, and temperature information as part of the free nerve ending system. He further suggests that in the taste bud system, the facial nerve, N. VII, carries information regarding sweet, bitter, salty, insipid, and sour stimuli, and that the glossopharyngeal nerve (IX) carries information regarding sweet, bitter, metallic and umami compounds. The role that the vagus nerve, N. X, has in taste perception is not well understood (Boudreau, 1986; Yanagisawa, 1998).
     Henning (1916) proposed a tetrahedron of the four basic tastes of salty, sweet, sour, and bitter, could define the taste realm if the basic tastes were placed at the corners of the tetrahedron. This work assumed that other taste sensations were nothing more than mixtures of primaries that were located on the edges or surfaces of the tetrahedron (Bartoshuk, 1978, Delwiche, 1996). However, studies on food have revealed that these four sensations alone are not sufficient to describe the perception of various foods (Boudreau et al, 1979; Boudreau, 1983). One may demonstrate that certain stimuli still fall outside Henning's (1916) tetrahedral model of the basic tastes by eliminating olfactory stimulation using a technique of forcing a gentle stream of air into the nostrils (Mozell et al, 1969). The same results may be achieved by using totally anosmic subjects (Mozell et al, 1969; Schiffman, 1980), demonstrating that the four basic tastes in conjunction with olfaction is inadequate to fully describe the flavor of foods. Anosmics, or those who have lost the ability to smell, often describe their food as tasteless or bland. Testing their perception of sweet, sour, bitter, and salty stimuli reveals that their ability to perceive these basic tastes has not been altered (Plattig, 1988). Although the flavor of the food constituents clearly has been modified, it is not the taste but rather the volatile aromas that migrate from the mouth through the choanae (posterior or pharyngeal nares) to the olfactory receptors that are not perceived in these cases (Piggott, 1988).
     Schiffman and Erickson (1971) attempted to expand the taste space through the use of multidimensional scaling. However, this work quickly met with criticism endemic to the use of multidimensional scaling (McBurney and Gent, 1979). They argued that multidimensional scaling did not lead to the discovery of new taste dimensions, but rather served to enhance understanding of the relationships among already established taste dimensions. They further argued that scaling of a small subset of stimuli often gives fundamentally different results than scaling the entire set of stimuli. As such, they argued studies using multidimensional scaling to date were incomplete and inadequate to properly model the complete taste space and results were often confounded by the specific stimuli selected for the study.
Tastes such as metallic, soapy, alkaline, and others, have been suggested, but their modes of perception are not well understood, and as such, they have not been universally accepted in the taste literature. Metallic taste is often reported following tastings of metal salts. A metallic sensation is also reported after tasting certain formed metals, though according to their surface electric potentials, these formed metals might actually produce what has been classified as an electric taste (Plattig, 1988). Cathodal DC stimulation has been shown to cause alkaline taste (soapy), while anodal stimulation leads to the perception of a sour taste (Plattig, 1988). This suggests that these tastes may both be linked to electric taste, though this leaves the exact nature of the metallic taste somewhat of a mystery. The reported tastes of metallic, soapy, or sour in carbonated beverages may also be due to altered ion balance within the taste buds, which may disrupt the normal pattern interactions of receptor cells (Roper, 1989), and lead to what are sometimes known as "phantom tastes," often associated with neurological damage or pathology (El-Deiry & McCabe, 1990). This contrasts with the research of Skramlik (1926), who reviewed work that suggested that the alkaline and metallic tastes result exclusively from olfactory stimuli and disappear when such stimuli are excluded.
     No studies to date have investigated the taste response of individual human nerve fibers; as such, no physiological system has been described as an analog for the human metallic sensation. Boudreau (1986) has suggested the PGsalt system of the rat petrosal ganglion as a possible analog since it responds to solutions with metallic cations without the Na+ and Li+ specificity shown by rat GGsalt units. Caution must be used in applying such analogs to the human taste system. Studies of basic physiological measurements reveal that notable difference exist among mammals (Pritchard, 1991). Comparisons of humans to great apes and monkeys reveal fundamental differences in central nervous system gustatory anatomies (Pritchard, 1991), and differences among different mammalian orders in anatomy is well documented (Smith-Swintosky et al, 1991). Even among members of the same species, there is evidence that taste transduction mechanisms for a specific tastant may have great variability (Kinnamon and Cummings, 1992). These findings beckon restraint when attempting to apply physiological models developed from animal experiments to humans (Delwiche, 1996).
     In human studies anesthetizing the chorda tympani branch of the facial nerve via the ear canal (Yanagisawa et al, 1992), 40% of subjects reported taste phantoms that appeared at the contralateral rear of the tongue when stimulated by non-tastant containing solutions. Phantoms most often were described as salty, sweet, sour, bitter, or metallic. By anesthetizing the location of the reported taste, one could eliminate the taste phantoms. Norgren and Pfaffman (1975) have shown that topical anesthesia eliminates spontaneous nerve activity, which led Bartoshuk (1993) to suggest that anesthetizing the chorda tympani allows the brain to perceive the basal activity of the glossopharyneal nerve, thus explaining the source of the phantom tastes.
     Bartoshuk (1993) discusses one subject who, recovering from a mastoidectomy that likely damaged or transected the chorda tympani, complained of a constant intense salty taste. Rinsing with a topical anesthetic comprised of 0.5 % dyclonine and 0.5% diphenhydramine in 0.9% saline led to a reduction of intense salty taste, presumably by eliminating the glossopharyngeal input. Early studies by Bartoshuk and Kveton (1991) suggest that neural responses from the chorda tympani nerve are usually inhibited by the neural responses of the glossopharyngeal nerve, but later research suggests that patients with localized damage in taste system often do not report a decrease in their ability to perceive tastes (Bartoshuk, 1993).

Perception of metallic attributes by metallic stimuli in non-food systems

     Metallic has been interchangeably used to describe taste, mouthfeel, and aroma, depending on the context. Gonázlez Viñas and Salvador (1998) compared method of limits ASTM E-679 for threshold determination (1991) and Lundahl's threshold determination method based on scales using iron (II) sulfate heptahydrate as a standard for metallic tastes in the range of 1.3 - 8.0 mg/L, as suggested by International Standard ISO 3972 (1991). Böröcz-Szabó (1980) notes that iron (II) sulfate, also known as ferrous sulfate or FeSO4, is the iron compound most advantageously utilized by humans. Using a set of 21 panelists, Gonázlez Viñas and Salvador showed judges were less able to correctly identify metallic tastes than other taste categories of sweet, salty, acid, bitter, and umami. This may be a remnant of panelist familiarity with these other taste qualities in contrast to their familiarity with metallic stimuli. Regardless, the authors showed using ASTM E-679 methodology that sensitivities for FeSO4 in water ranged from 2.26 - 9.56 mg/L, with a group mean threshold that was 7.90 mg/L. This was contrasted with Lundahl's method, which found all individual and group detection thresholds to exceed 8.00 g/L. They compared this with the suggested ISO range of 1.3 - 8.0 mg/L, arguing that the international standard ISO 3972 concentrations were lower than they should be for use of Lundahl's method with untrained panels. This may be due to the fact that, as Böröcz-Szabó (1980) notes, in tap water, both ferrous and ferric salts form red precipitates soon after dissolution, though in deionized water, only ferric salts produced a red precipitate. Ferric salts of FeCl3 were colorless to concentrations of 1 mg dm-3, while Fe2(SO4)3 solutions were colorless to 2 mg dm-3.
     Metallic taste was reported as a major defect in the water supplied by water utilities, based on a taste and odor questionnaire with 377 responses from the United States and Canada (Suffet et al, 1996). The authors used a taste and odor wheel (Mallevialle and Suffet, 1987) that classified metallic as a "mouthfeel" sensation rather than a taste. Although sour water was the most reported taste problem, metallic taste was the second most frequently reported taste problem, followed by chlorination. Linssen et al (1991) applied factor analysis of tainted water in pouches of aluminum lined with low-density polyethylene to data from a panel of 48 assessors. Results reveal that metallic ratings were significantly different from control experiments, suggesting the pouches influenced the flavor of the water.
     The work of Nakagawa (1972) may justify the high incidence of reported off-metallic flavor in water sources (Japanese article - getting translated - insert interpreted regarding specific thresholds here). They showed that mutlivalent metallic salts were shown to form soluble complexes with proteins in water that lead to extreme levels of astringency. Presumably, if these multivalent compounds were present in the water, they could bind with salivary proteins and lead to the reported off-metallic flavors noted in the work of Suffet et al (1996) and Linssen et al (1991).
     Research by Roy (1992) shows that with saccharins, the usual bitter, metallic taste is removed when a carbon atom is strategically replaced by a sulfur or an oxygen atom. This may be significant when one considers the AH-B-X model of sweetness reception, though the implications of this fact are not fully understood.
     Exposure to heavy metals and toxins has also been shown to lead to metallic tastes. Reports (FDA Web Document B, 2002) discuss metallic taste as a symptom of copper and lead poisoning with latency periods of less than one hour, as well as a symptom of ciguatera poisoning with a 1-6 hour latency period. Barceloux (1999) reports zinc oxide as the most common cause of metal fume fever, of which metallic taste is a symptom. This study reports that seafood, meat, dairy products, nuts, legumes, and whole grains contain relatively high concentrations of zinc. Mahaffey et al (1975) reports that meats, poultry, and fish are considered to contain an average of about 25 mg Zn/kg.
     Studies (Agency for Toxic Substances and Disease Registry, 1997) show the ambient concentration of zinc in the air is generally less than 1 microgram per cubic meter. Zinc chloride, zinc oxide, zinc stearate, and zinc sulfate are commonly used food additive zinc salts that have legally have generally recognized as safe, or GRAS, status. Zinc does not accumulate in the body, with 70-80% eliminated fecally, and 15-25% eliminated through sweat and urine. Zinc oxide fumes lead to the onset of metal fume fever within 4-10 hours of exposure, though metal fume fever may also occur following exposure to aluminum, antimony, copper, iron, magnesium, manganese, and nickel fumes.
     Studies (International Labour Office, 1983) show that occupational long-term exposure in steel workers who handle tellurium pellets often leads to reports of metallic tastes, as well as oft reported accounts of strong garlic odors. Müller et al (1989) reports a study of the toxic effects of tellurium reported for a 37 year old non-occupationally exposed woman with tellurium intoxication. She experienced nausea and vomiting in conjunction with a metallic taste in the mouth and a sharp garlic odor in her breath, sweat, and excrement.
     Kachru et al (1989) report chronic lead poisoning among nine adults and nine children attributable to lead fumes and lead oxide dust produced during the extraction of gold and silver from jewelers waste. Pre-examination interviews showed five of seven male silver jewelery workers reported sweet metallic tastes in their mouths. Heamatological examination and urinalysis showed significant increases in blood lead, zinc protoporphyrin, urinary excretion of lead and aminolevulinic acid, with associated decreases in blood aminolevulinc acid dehydratase activity.

Perception of metallic attributes in non-animal flesh food systems

     Yau and Huang (2000) conducted sensory tests on membrane processed water used in Oolong tea using descriptive analysis and preference tests among 15 trained panelists. This experiment used 12 g Oolong tea stem extracted by 600g water at 100o C for 5 minutes to serve as the metallic aroma standard. This same sample, when stirred for 20h with a magnetic stir bar, also served as an oxidized aroma standard. The stems of the Oolong tea before extraction served as the roasty aroma standard. It is unclear how the standards were justified for this experiment. Regardless, when using unprocessed, nanofiltrated, and reverse osmosis treated water, both oxidized and metallic aromas showed significant increases with each of the respective treatments. The unprocessed tea sample, showing the lowest scores for oxidized and metallic aromas, showed the lowest hedonic scores among the samples. This is a rare case where the presence of a metallic note does not inversely correlate with hedonic liking. However, this study does not consider metallic aromas a key attribute of aroma; PCA of the data considers metallic aroma as one of the least influential aspects of the descriptive analysis, while oxidized aroma is a descriptive attribute noted in PC1. The authors postulate that the higher presence of metallic aromas in the reverse osmosis tea samples compared to those using nanofiltrated water may be due to the presence of monovalent ions, though this blanket consideration is applied to many attributes beyond metallic aroma and is not fully investigated.
     In contrast to the prior study, Kumasawa and Masuda (1999) used aroma extract dilution analysis to detect odorants of the volatile fraction of a Japanese green tea (Sen-cha) to find odor-active peaks by GC/MS and/or multidimensional GC/MS. They report (Z)-1,5-octadien-3-one as having one of the highest flavor dilution values, FD = 1000, justifying it as one of the most important odorants of Sen-cha. Guth and Grosh (1993) in prior work had suggested that this was one of the 3 most important odorants in this tea, which differs in production methods from Chinese green tea. This compound is noted in both studies as having a characteristic metallic odor.
      There exists evidence suggesting that iron content can negatively affect the taste of teas. Chakraborty and Baruah (1971) reported that excess iron in water produces a "brassy, harsh and metallic taste" in cup-infusion tea, arguing that iron content of more than 2 ppm is detrimental to tea quality. Unfortunately, no correlations were shown in the work of Guth and Grosh (1993), Kumasawa and Masuda (1999), or Yau and Huang (2000) relating the indirect effects of iron content to the levels of metallic odorants in tea.
     During informal discussions at the Eighth International Colloquium on the Chemistry of Coffee in 1977, two experienced coffee tasters reported that black coffee brews occasionally had a rather unpleasant astringent taste that was often accompanied by a metallic taste that was quite distinct from bitterness. Clifford and Ohiokpehai (1983) have explored these reports in follow-up studies that focused on preliminary organoleptic evaluation of dicaffeoylquinic acid, chosen for its similarity to galloylquinic acids used in astringency research. Results suggest that this compound cannot be adequately be described by the term "astringent," though it does present an easily detected and lingering metallic taste in water and coffee in the range of 50-100 mg/L.
     Böröcz-Szabó (1980) performed a broad study that classified foodstuffs based on the concentrations at which ferrous iron concentrations affected perception in the log10 schemes of those requiring concentrations of 1000, 100, and 10 mg/L to be detected. The first group contains fruit juices, the second group contains wines and liqueur, and the third and most sensitive group contains beer and milk. At high concentrations in fruit juices, bitter tastes were reported in addition to metallic flavor, and salty notes were reported above 1500 mg/L. In wines, ferrous salts were postulated to liberate hydrogen sulfide, the unpleasant smell of which likely masked the flavor effects of the iron salts and minimized the identification of metallic notes. Both ferric and ferrous salts caused substantial changes in beer, both in loss of CO2 and in increases in turbidity. For milk products, both ferric and ferrous salts imparted not only a metallic taste, but also led to panelist reporting that the milk smelled stomach turning or rancid. Some individuals reported the spiked milk samples smelled or tasted paper-like. The counter-anions present in these studies had little effect on sensory impact. Böröcz-Szabó notes that the taste of deionized water is unpleasant, which makes it difficult to distinguish it from ferrous solutions because of the inherent difficulty distinguishing between two unpleasant tastes. Reports from panelists reveal lingering astringent, metallic sensations remained in their mouth for a long time after testing. One may hypothesize that the presence of lipids milk and iso-alpha-acids in beer leave them particularly prone to degradation by both ferric and ferrous salts, while fruit juices, which are high in antioxidant potential, can mask such contaminants to a much greater extent.
     Gains and Thomson (1990) employed free-choice profiling using 19 consumers testing 17 canned lagers and analyzed their results by general Procrustes analysis. Their results show that PC1 was responsible for 43% of the variance and contrasted thin mouthfeel to rich, full, viscous, syrupy, thick mouthfeel, sweet, syrupy flavor, strong, and alcoholic flavor. PC2 was responsible for 8% of the variance and contrasted good head retention, bitter flavor, and yeasty malty flavor against long aftertaste, metallic, sour taste, and syrupy mouthfeel. The authors suggest that most important discriminator of different beers and lagers are mouthfeel terms, while in this study metallic notes were not major contributors to flavor.
     Gautschi and Schmid (1997) report using a panel of 5-7 experienced creative panelists that the diketopiperazine (DKP) cyclo(Val-Pro) found in beer has a "lingering, metallic, salty taste at 10 ppm" and found that increased levels of DKP's correlated with darker levels of malt. De Graaf et al (1990) found that when diketopiperazines are combined with theobromine, they result in a pronounced bitter metallic taste. Bonvehí and Coll (2000) extended these findings to DKP levels in roasted cocoa powder, showing that a trained panel of 10 judges considered levels of 80 mg kg-1 unacceptable under all roasting conditions tested. At roasting temperatures of 145o C for 3 or 10 min, judges gauged the product as unacceptable, at levels as low as 74.2 DKP mg kg-1 cocoa powder, whereas the product was gauged as acceptable at both 125 and 135o C for the same roasting times with DKP levels not exceeding 72.3 mg DKP kg-1 cocoa powder. Furthermore, adding reducing sugars to the cocoa powder lead to reduced levels of DKP in conditions that had previously yielded unsatisfactory levels, and supporting the claim of increased levels of DKPs with more intense kilning regimes.
     Lindner et al (1997) evaluated a dihydrochalcone sweetener, CH-401 in foods. When added to chocolate and vanilla parfait (100 mL milk, 4 egg yolks, 500 mL sweet cream), judges rated samples less well in hedonic tests. The development of metallic after-taste was reported as the source of the decrease in hedonic ratings. Subjects also reported a metallic aftertaste in yellow cream (500 mL milk, 35 g flour, 15 g cream-powder, 3 eggs), and both cooked and cold buttercreams were identified as having strong metallic tastes. The CH401 content in these fives samples was 12.20, 12.0, 4.5, 102.6, and 102.6 mg per 100 g of samples, respectively. The authors report that in the case of products containing egg-yolk and fats, the sweetening capacity is slight, but the resulting metallic taste is distinct.
     Other compounds have been shown to correlate to metallic taste. Blouin and Vinnet (1985) report that phospholipid portion of cottonseed flour is responsible for metallic taste. DuPont et al (1994) observed a correlation between bitter and metallic and between these tastes and total alkaloid content of slurries of milled lupin seed in water. Metallic tastes were considered weak tastes in these samples by these judges, when compared to other tastes that were evaluated. Bitter and green, and beany tastes were the strongest tastes. No relationship was observed with tannin content and metallic taste.
     Poretta (1985) reports that by PCA, citric acid is most closely related to metallic taste among compounds examined for canned whole tomatoes. However, this study only tested various acids and sugars, as well as pectins, protopectins, pectic acid, ethanol, and NaCl. No analysis of any metallic compounds, or any compounds believed to elicit a metallic flavor, was explicitly performed. Mahadeviah et al (1983) report that both soldered and welded cans packed with different products retained good vacuums, but organoleptic evaluations of potatoes stored in brine at 37o C report metallic tastes.
     Davies (1936) found that iron present in milk in the form of lactacte expressed a metallic flavor at concentrations of 15 ppm. Reports from El Dadek et al (1972) suggest that in pickled kareish cheese, low quality cheeses are often reported as having firm texture or sharp acidic tastes that are usually accompanied by metallic flavors. No sensory analysis was reported in these studies to back such claims. Degouy (1993) showed by using Simplesse 100, a whey by-product of Gouda manufacturing, a low-fat cheese was made and subjected to an eight person sensory panel. It was reported that after 15 weeks, no metallic or bitter flavors occurred. Work by Rice and McMahon (1998) compared control Mozzarella cheese to iron fortified Mozzarella cheese and found that fortification with 25 or 50 mg of either FeCl3, iron chelated to whey protein, or iron chelated to casein proteins, per kg of cheese; showed increases in oxidized, metallic, and off-flavors. Judges who were trained to detect oxidized and metallic flavors were more sensitive than the TBA assay in detecting differences in oxidation flavors in cheese, suggesting that the TBA assay does not sufficiently model these flavors in cheese. The authors suggest that oxidative and metallic flavors of the iron fortified cheeses were less than expected, considering the role of iron in lipid perodixation. When the Mozzarella cheeses were cooked on pizzas, no differences were perceptible by the consumer panel.
     Heiler and Schieberle (1997) state that it is well known in the dairy industry that due to rapid off-flavor development, the shelf-life of sour cream buttermilk is much shorter than that of fermented buttermilk. The typical off-note developed during storage is often described as metallic (Frank, 1984). Sour cream buttermilk contains 40% fat before fermentation, whereas the so-called fermented buttermilk is usually derived from sweet-cream buttermilk of 1% milkfat that is treated with butter starter cultures after separation of butterfat. Using comparative aroma extraction dilution analysis applied to flavor extracts prepared from fresh fermented sweet-cream buttermilk and a stored sour-cream buttermilk exhibiting intense metallic flavors, Heiler and Schieberle (1997) noted that metallic smelling odorants (E,Z)-2,6-nonadienol and 4,5-epoxy-(E)-2-decenal, as well as the mothball smelling odorant 3-methylindole were the most important odorant contributors to metallic defects in buttermilk. The (E,Z)-2,6-nonadienol was reported as the key odorant responsible, which reportedly has been found in several plant materials. Hatanaka et al (1975) suggest that this compound may be formed in cucumbers from an enzymic reaction starting from alpha-linolenic acid as a precursor.
Buettner and Schieberle (2001) quantified odor active compounds in fresh, hand squeezed White Marsh seedless grapefruit juice using stable isotope dilution assays and calculation of odor activity values. In this study, they showed that trans-4,5-epoxy-(E)-dec-2-enal, noted as offering a metallic odor in the previously mentioned studies, was among the most potent odorants. Prior work by Buettner and Schieberle (1999) had identified the presence and nature of odorant using GC/Olfactometry of serial dilutions, or Aroma Extract Dilution Analysis. These results found concentration in fresh hand-squeezed grapefruit juice to be 0.0031 mg/kg. Rychlik et al (1998) suggest odor threshold of these compounds in air is in the range of 0.0006-0.0025 bg/L, revealing the extreme olfactory sensitivity in humans to this compound. Shrader and Johnson (1934) report that as little as 5 ppm iron or copper can cause off-flavors in orange juice, though this is not clearly identified as a metallic defect.
     Trans-4,5-epoxy-(E)-nonenal and trans-4,5-epoxy-(E)-2-decenal appear again in aroma studies by Stephan and Steinhart (1999). By performing aroma extract dilution and GC-sniff techniques, these compounds are reported as metallic contributors in standardized, hydrolyzed, and oil free commercial soybean lecithins. Modified CHARM (combined hedonic and response measurements) analysis rated trans-4,5-epoxy-(E)-decenal with a high CHARM values of 701, 907, and 649 for standardized, hydrolyzed, and oil free commercial soybean lecithins, whereas the nonenal form rated CHARM values as <3 in all three samples, showing lower CHARM values than any other 61 odorants studied. Flavor dilution values for the decenal forms were 1024, 1024, and 512 versus values of 32, 16, and 4 for the nonenal for standardized, hydrolyzed, and oil free soybean lecithins. Metallic is rated as a taste in the actual studies, but regarded in this study by the authors as a retronasal sensation that increases through hydrolysis of lecithins. The sensation decreases through acetone deoiling.
     Volatile flavor compounds of miniature beefsteakplants (Mosla dianthere Maxim) were analyzed through GC/MS/O (Kim et al, 2001). The results report that (Z)-3-hexenol is an important aroma-active compound, with a grassy/leafy/metallic aroma. In chemical contrast, Buettner and Schieberle (2001) report (Z)-3-hexanal as important to the aroma of grapefruit juice, offering only green/grassy notes. The concentration of (Z)-3-hexenol in miniature beefsteakplants is reported as 4.2 ppm with an aroma threshold in water reported by Buttery et al (1971) of 0.070 ppm. Hatakanaka (1996) reports this to be biosynthesized in green leaves via lipoxygenase and hydroperoxide lyase-mediated lipid oxidation.

Perception of metallic attributes in animal flesh food systems

     Extensive work has been performed analyzing metallic notes in animal flesh foods, as well in the previously mentioned non-animal flesh foods. Seafood, chicken, pork, and beef have all identified metallic off-flavors in their products. In each case, this is considered a defect. Refsgaard et al (1998) note that the most pronounced sensory changes during frozen storage of salmon are increased intensity of train oil taste, bitterness, and metal taste, which result not from volatile oxidation products, but instead from compounds of low volatility. Increased levels of free fatty acid during frozen storage are reported for several species of fish (Shewfelt, 1981). Refsgaard et al (2000) demonstrated that polyunsaturated acids added to fresh salmon mince at levels found in frozen stored salmon induced similar sensory phenomena as those found in salmon after frozen storage. This suggests that enzymatic hydrolysis of neutral lipids plays key role in sensory deterioration of salmon in frozen storage. Taste assessment studies showed that palmitoleic acid, linoleic acid, eicosapentaeonic acid, and docosahexaenoic acid show high intensity of train oil, bitter, and metal taste. It is interesting to note, however, that the addition of the free fatty acids did not affect the odor of the fish. They suggest that only small amounts of oxidation products from the fatty acids are formed after addition of the fatty acids and chopping of the salmon meat. As such, the total concentrations of such by-products do not exceed the odor thresholds of these products in meat model systems.
     Lovell (1983) described major off-flavors in fish and included the term metallic. A descriptive analysis study performed by Johnsen et al (1987) using 18 people experienced in tasting catfish rated metallic as an important flavor phenomena, but listed metallic as the chemical feeling factor on the tongue, rather than a taste, and described it as a "flat" sensation associated with iron and copper solutions. Occasional reports of metallic flavored seafood cocktails have been given from time to time. These may result from biochemical reactions, microbial spoilage, or accidental contamination by manmade products. Whitfield et al (1982) have shown that the garlic-metallic off-flavors in the royal red prawn (Hymenopenaeus sibogae) are likely due to bis-(methylthio)-methane, oct-1-en-3-ol, and (5Z)-octa-1,5-dien-3-ol. Studies show that these off-flavor compounds originate in the heads of royal red prawn, but upon cooking the compounds diffuse into the edible portions of the meat (Davidson, 1984).  Hanson and Duckworth (1982) report that when pure carbon dioxide is used in controlled atmosphere packaging on fish, marked drip loss occurs, accompanied by the development of a metallic/acidic taste. The same is not reported to occur in red meat packaging.
     Metallic has long been described as an immediate off-taste/odor developed after irradiation (Morgan and Siu, 1957). Studies by Risvik (1986) using descriptive analysis of irradiated beef and bacon were performed using 10 trained panelists. Unfortunately, no chemical analysis was performed to determine which components form the characteristic metal tastes. Other than reporting samples as tasting metallic, subjects reported a prickling sensation on the tongue that lasted for several hours after assessment. In Risvik's study, metallic taste increases correlated with decreased juiciness, color saturation, and increased off-taste, sweetness, and rancidity. Other studies on beef have listed metallic as a descriptor (Berry et al, 1980; MacLeod and Coppock, 1978; Persson et al, 1973).
Buscailon et al (1994) studied hams processed by dry curing for either 179 or 273 days and submitted them to chemical and sensory analysis. A 12 member trained panel rated intensity and persistence of metallic aroma, again defined as the result of retronasal olfactory perception. No differences were noted in between the aged samples for total intensity or persistence of metal aroma, and the authors fail to identify the compounds responsible for the source of metallic aromas. A chemical analysis using GC/MS is presented, but reports many of the compounds as unidentified or unknown.
     Jeremiah et al (1996) analyzed 80 cured and smoked hams for fat, sodium, potassium, chloride, and nitrite content. Hams were subjected to a highly trained flavor profile panel. Each flavor note was reportedly described along with the order of appearance of each note. In this study, fat content was correlated to metallic aftertaste. The authors fail to identify specific compounds responsible for the metallic aftertaste.
     Heymann et al (1990) evaluated pork roasts cooked to various endpoints by sensory and chemical analyses. A trained ten person panel evaluated samples and specifically rated metallic flavor, defined as "The degree of metallic 'sensation' perceived." Increased endpoint temperatures were reported as associated with decreased metallic flavor. Canonical variate analysis correlated increased endpoint temperatures with increased brown color and decreased pink color, likely related to the conversion of oxymyoglobin, metmyoglobin, and myoglobin to denatured metmyoglobin or loss of nitrosylhemochrome. This suggests that for pork roasts, the state of myoglobin may influence metallic taste. Additionally, juiciness is reported to decrease with increasing endpoint temperatures. Therefore, it is possible that moisture content of meats plays a role in perception of metallic tastes. However, juiciness is shown as orthogonal to metallic taste, reducing the likelihood of a meaningful relationship between these attributes. No attempts were reported to isolate the chemical compounds responsible for the metallic taste.
     A ten member panel was used to create a terminology list useful for the analysis of cooked chicken patties (Lyon, 1987). A PCA was performed on the terms generated by the panelists over several sessions via free word association. Metallic was listed as one of the descriptors. Factor analysis of taste data formed 8 groups of terms, and grouped metallic with terms like gamey/fowl-like, canned, chemical, and astringent; the last two terms had low factor loadings and low frequencies of use. Subsequent factor analysis of the terms remaining after eliminating those with low factor loadings and low frequencies created 6 factors and this time grouped metallic with liver/organy, gamey/fowl-like, and canned. Here, canned had low loadings, so it was dismissed as a useful term. Eventually, the 45 word list initially developed was reduced to 12 terms. Metallic was the only term not associated with taste or smell, but instead defined as a feeling factor on the tongue.

Perception of metallic attributes resulting from medication
  
   Research has shown that drugs, including antiretroviral drugs, are secreted into saliva (Mucklow et al, 1978) at levels high enough to exert adverse effects on taste sensation by either modifying taste transduction mechanisms or producing a taste of their own. Many drugs are converted to active metabolites by intracellular enzymes, but their volumes of distributions are high, which suggests that the unmetabolized drug is dispersed into extravascular spaces (Physicians Desk Ref, 1997).
In a study of eighteen adult male outpatients with AIDS Glover et al (1995), sixteen patients reported a metallic taste resulting from intravenous administration of pentamidine isethionate. This taste arises independent of eating, but may be masked or magnified during eating. Following administration, pentamidine isethionate is distributed throughout the body with a mean elimination half-life of 29 hours in patients with normal renal function. Sweet foods were reported to make the unpleasant taste better, while milk and tap water were reported to make the taste worse. Milk shakes and ice cream, being sweet, were reported to make the unpleasant tastes decrease. Metallic taste persisted 24 hours a day from day 6-21, at times leaving patients hungry but unable to eat.
     Ritonavir acts as an inhibitor of HIV protease with activity against the Human Immunodeficiency Virus, HIV, and is also reported to have a bitter metallic taste (FDA Web Document G, 2000). In contrast to pentamidine isethionate, ritonavir is practically insoluble in water. Didanosine and lamivudine are nucleoside analogs also used to treat HIV. Didanosine is described primarily as bitter, metallic, sour, and medicinal, while lamivudine is described as bitter, metallic, medicinal, astringent, sour (Schiffman, 1999).
     Histamine dihydrochloride is used with interleukin-2 to treat adults with advanced metastatic melanoma that has metastasized to the liver. Metallic taste is given as a common adverse reaction to taking this medication (FDA Web Document I, 2000). Amifostine is a thiol compound with wide cytoprotective activity. Genvresse et al (2001) evaluated amifostine use in the elderly. This study compared side effects related to doses of 740 mg m-2 in combination with chemotherapy given to patients 70 years and older, as well as with younger patients to determine if age has an affect on the tolerability of amifostine. Between 4-5% of the patients reported metallic dry mouth as a side effect of the drug.
     Ivancev et al (1989) studied the effect of an iodinated lipid emulsion in 16 patients for computed tomography of the liver. Fifteen of the sixteen patients suffered from malignant lesions, while only one suffered from focal nodular disorder. Four of the patients, suffering from malignant lesions, described a metallic off-taste as moderate and lasting only ten minutes. The only patient tested with focal nodular hyperplasia reported the metallic taste as well, and described it as disturbing and lasting beyond the end of the 24-hour study. This patient had uncharacteristically low iodinated lipid emulsion uptake into the liver and high uptake into the spleen when compared to the 15 patients who suffered from malignant lesions. Iodine containing compounds have been shown to cause metallic tastes in other contexts. Potassium iodide can be used to help prevent radioactive iodine uptake in the event of a radiation emergency (FDA Web Document D, 2001).
      Gelenberg et al (1989) report studies comparing standard and low serum levels of lithium dosages for treatment of bipolar disorder. They show that standard doses, adjusted to achieve serum lithium concentrations of 0.8 - 1.0 mmol per liter, were more effective at treating bipolar disorder than low doses, adjusted to achieve blood serum lithium concentrations of 0.4 -0.6 mmol per liter. However, side effects were reported more frequently in the standard dose group than in the low dose group. Interestingly, metallic taste was reported only in the standard dose group, with roughly 20% of patients receiving therapy reporting a persistent metallic taste after six months of continuous administration.
    Metronidazole is the most extensively studied antibacterial treatment for inflammatory bowel disease, is quickly absorbed and distributed throughout the body. MetroLotion is a metronidazole containing lotion that acts as an antiprotozoal and antibacterial agent (FDA Web Document F, 1998). A metallic taste is one of the most commonly reported adverse effects with topical use of metronidazole. Aminosalicylates, corticosteroids, and immunomodulators also used to treat inflammatory bowel disease report adverse effects, but the presence of a metallic taste is not listed as a side effect of these treatements (Stein and Hanauer, 2000). Reports exist showing that metallic taste has occasionally been reported after intravenous administration of higher doses of clindamycin phosphate, a drug used in the treatment of infections caused by anaerobic bacteria (FDA Web Document C, 1998).
     Another topical agent that causes a metallic taste as an adverse reaction is Clinoril. This is a non-steroidal anti-inflammatory drug that also has utility as an analgesic and antipyretic. (FDA Web Document J, 2001). This drug is used for many arthritic and inflamed muscle conditions.
Metformin decreases the absorption of vitamin B12 and folic acid. Melchior and Jaber (1996) discuss the utility of metformin as an antihyperglycemic agent to treat type II diabetes. In the absence of insulin, metformin has no effect. Metallic taste is one of the most frequently reported side effects of metformin used in this context. Glucophage tablets are another form of metformin, composed of metformin hydrochloride, used in the treatment of type II diabetes and is freely soluble in water. There are reports among about 3% of users that these tablets have an unpleasant metallic taste that lasts for a short time when starting this medication (FDA Web Document H, 2001).
     Early symptoms of excessive levels of blood calcium include a metallic taste in the mouth. Hectoral, used to treat lower high parathyroid hormone levels in patients undergoing chronic kidney dialysis, affects calcium absorption in the body (FDA Web Document A, 2002). Rocaltrol is a synthetic vitamin D analog that helps regulate absorption of calcium in the gastrointestinal tract (FDA Web Document E, 1998). Calcitriol occurs naturally in humans, 99.9% bound in blood. Excessive dosages of Rocaltrol can lead to hypercalcemia, leading to a characteristic metallic taste. The symptoms of Rocaltrol are similar to those encountered in cases of vitamin D intoxication.
     An interesting interaction has been shown to occur for smokers using silver actetate as a deterrent to smoking. The silver interacts with smoke from cigarettes to produce an adverse metallic taste that helps aid in reducing the desire to smoke. Malcolm et al (1986) report evidence of silver acetate solutions being used as a deterrent to smoking since before 1940. They performed a study with one thousand subjects given either silver acetate gum or a placebo to help act as a deterrent to smoking. Statistical differences existed between the control and experimental conditions. Hymowitz and Eckholdt (1996) later studied the effect of silver acetate in lozenge form. Five hundred adult smokers were assigned to either receive a 2.5 mg silver acetate lozenge or a placebo. However, they report that some smokers did not consider the reaction to be very aversive, whereas some others considered the reaction to be very aversive, suggesting different sensitivities may exist among people.

References:

Agency for Toxic Substances and Disease Registry, US Public Health Services. (1997) ATSDR's Toxicological Profiles: Zinc. Boca Raton, FL: Lewis Publishers, CRC Press.

ASTM (1991) Standard Practice E679. Determination of odor and taste thresholds by a forced choice ascending concentration series method of limits. American Society for Testing materials, Philadelphia.

Barceloux, D.G. (1999) Zinc. Journal of Toxicology and Clinical Toxicology, 37 (2): 279-92.

Bartoshuk, L. M. (1978) "History of taste research" in Handbook of Perception (Carterette, E.C., Friedman, M. P., eds), Academic Press, pp. 3-18.

Bartoshuk, L.M. (1993) The biological basis of food perception and acceptance. Food Quality and Preference. 4, 21-32.

Bartoshuk, L.M., Kveton, J. (1991) Peripheral source of taste phantom demonstrated by topical anesthesia. Chemical Senses, 16, 499-500.

Berry, B.W., Maga, J.A., Calkins, C.R., Wells, L.H., Carpenter, A.L., Cross, H.R. (1980) Flavor profile analysis of cooked beef loin steaks. Journal of Food Science, 45: 1113-1115, 1121.

Blouin, F.A., Vinnet, C.H. (1985) Taste/flavor characteristics of edible cottonseed flour. Paper No. 85, Abstracts 190th American Chemical Society National Meeting, Sept 8-13. Dib. Agricultural and Food Chemistry, Chicago, IL.

Bonvehí, J.S., Coll, F.V. (2000) Evaluation of purine alkaloids and diketopiperazines contents in processed cocoa powder. Euro Food Res. Techno, 210, 189-195.

Böröcz-Szabó, M. (1980) The influence of iron contamination on the sensory properties of liquid foods. Acta alimentaria, 9 (4) 341-356.

Boudreau, J. C. (1986) Neurophysiology and Human Taste Sensations. Journal of Sensory Studies. 1, 185-202.

Boudreau, J. C., Oravec, J., Hoang, N.K., White, T.D. (1979) Taste and the taste of foods. Ch. 1 in Food Taste Chemistry (J.C. Boudreau, ed.) American Chemical Society, Washington, D.C., p. 1.

Boudreau, J.C. (1983) Chemical senses and food flavor: an overview. Ch. 2 in Instrumental Analysis of Foods Vol 1, G. Charlambout and G. Ingeltt, eds, Academic Press, New York, p. 17.

Brand, J. G., Bryant, B.P. (1994) Receptor mechanisms for flavour stimuli. Food Quality and Preference, 5 (1/2) 31-40

Buettner, A., Schieberle, P. (1999) Characterization of the most odor-active volatiles in fresh, hand-squeezed juice of grapefruit. J. Agric. Food Chem, 47: 5189-5193.

Buettner, A., Schieberle, P. (2001) Evaluation of key aroma compounds in hand-squeezed grapefruit juice (Citrus paradisi Macfayden) by quantitation and flavor reconstruction experiments. J. Agric. Food Chem., 49: 1358-1363.

Buscailhon, S., Berdagué, J.L., Bousset, J., Cornet, M., Gandemer, G., Touraille, C., Monin, G. (1994) Relations between compositional traits and sensory qualities of French dry-cured ham. Meat Science, 37: 229-243.

Buttery, R.G., Seifert, R.M., Guadagni, D.G., Ling, L.C. (1971) Characterization of additional volatile components of tomato. J. Agric. Food Chem., 19: 524-529.

Chakraborty, S.; Baruah, A.C. (1971) Kind of water used in tea factories and its effect on quality. Two and a Bud, 18 (1): 25.

Clifford, M.N.; Ohiokpehai, O. (1983) Coffee astringency. Analytical Proceedings, 20 (2): 83-86.

Davidson, S. (1984) Analysing off-flavors. Rural Research, 122:11-14.

Davies, W.L. (1936) The Chemistry of Milk. Chapman and Hall, Ltd. London. P. 387.

De Graaf, C., Legger, A., Roozen, J.P. (1990) J. Sci. Food. Agric. 50, 363-375.

Degouy, E. (1993) The low fat cheese challenge. Dairy Industries International, 58 (10): 41-43.

Delwiche, J. (1996) "Are there 'basic' tastes?" Trends in Food Science and Technology, 7, 411-415.

DuPont, M.S., Muzquiz, M., Estella, I., Genwick, G.R., Price, K.R. (1994) Relationship between sensory properties of lupin seed with alkaloid and tannin content. J. Sci. Food Agric., 65: 95-100.

El Dadek, G.; Moneib, A.; Gharib, S.Y. (1972) Studies on improvement of Kareish cheese. II. Effect of salt and hydrogen ion concentration on the physical and chemical properties of the cheese. Agricultural Research Review, 50 (3): 179-184.

El-Deiry, A., McCabe, B.F. (1990) Temporal lobe tumor manifested by localized dysgeusia. Ann. Otol. Rhinol. Laryngol. 99, 586-7.

FDA - Food and Drug Administration, Web Document A (1999) Hectoral. (Updated 4/7/00 Revised: 3/15/01) www.fda.gov/cder/consumerinfo/druginfo/hectorol.htm, last accessed April 2002.

FDA - Food and Drug Administration, Web Document B (2002) Classification of illnesses attributable to foods. :www.fda.gov/ora/inspect_ref/iom/exhibits/x910c.pdf (last accessed: April 2002).

FDA - Food and Drug Administration, Web Document C (1998) Guidance for Industry: Labeling guidance for clindamycin phosphate injection, USP. www.fda.gov/cder/guidance/clindaiv.pdf (last accessed: April 2002).

FDA - Food and Drug Administration, Web Document D (2001) Potassium Iodide Recip Product Label. www.fda.gov/ohrms/dockets/dailys/01/Aug01/082801/sup0001.pdf (last accessed: April 2002)

FDA - Food and Drug Administration, Web Document E (1998) Rocaltrol www.fda.gov/cder/foi/label/1998/21068lbl.pdf (last accessed: April 2002).

FDA - Food and Drug Administration, Web Document F (1998) Physician Package Insert, MetroLotion. www.fda.gov/cder/foi/label/1998/20901lbl.pdf (last accessed: April 2002).

FDA - Food and Drug Administration, Web Document G (2000) Novir. www.fda.gov/cder/foi/label/2001/20659s26lbl.pdf (last accessed: April 2002).

FDA - Food and Drug Administration, Web Document H (2001) Clinoril. www.fda.gov/cder/foi/label/2001/17911s66lbl.pdf (last accessed: April 2002)

FDA - Food and Drug Administration, Web Document H (2001) Glucophage. www.fda.gov/cder/foi/label/2001/20357s22lbl.pdf (last accessed: April 2002).

FDA - Food and Drug Administration, Web Document I (2000) Histamine dihydrochloride www.fda.gov/ohrms/dockets/ac/00/backgrd/3671b1_07.pdf (last accessed: April 2002).

Frank, J. (1984) Improving the flavor of cultured buttermilk. Cultured Dairy Products Journal, 00, 6-9.

Gains, N., Thomson, D.M.H. (1990) Sensory profiling of canned lager beers using consumers in their own homes. Food Quality and Preference, 2, 39-47.

Gautschi, M., Schmid, J.P., Peppar, T.L., Ryan, T.P., Tuorto, R.M., Yang, X. (1997) Chemical characterization of diketopiperazines in beer. J. Agric. Food Chem., 45: 3183-3189.

Gelenberg, A.J.; Kane, J.; Keller, M.B.; Lavori, P; Rosenbaum, J.F.; Cole, K.; Lavelle, J. (1989) Comparison of Standard and Low Serum Levels of Lithium for Maintenance Treatment of Bipolar Disorder. The New England Journal of Medicine, 321 (22): 1489-93.

Genvresse, I.; Lange, C.; Schanz, J.; Schweigert, M.; Harder, H.; Possinger, K.; Spath-Schwalbe, E. (2001) Tolerability of the cytoprotective agent amifostine in elderly patients receiving chemothreapy: a comparative study. Anti-Cancer Drugs, 12 (4): 345-9.

Glover, J.; Dibble, S.; Miaskowski, C.; Geibert, R. (1995) Changes in taste associated with intravenous administration of pentamidine. Journal of the Associated Nurses of AIDS Care, 6 (1): 43-8.

Gonázlez Viñas, M.A., Salvador, M.D. (1998) Comparison of two simple methods for the measurement of detection thresholds for basic, umami, and metallic tastes. J. Sensory Studies. 13, 299-314.

Guth, H., Grosch, W. (1989) 3-Methylnonane-2-4-dione - An intense odor compound formed during flavour reversion of soya-bean oil. Fat Sciences Technologies, 93:225-230.

Hanson, N.A.; Duckworth, H.A.G. (1982) CAP = preservation. Packaging, 53 (625): 33.

Hatakanaka, A. (1996) The fresh green odor emitted by plants. Food Rev. Int., 12: 303-350.

Hatanaka, A., Kajiwara, T., Harada, T. (1975) Biosynthetic pathway of cucumber alcohol: transt-2-cis-6-nonadienol via cis-3,cis-6-nonadienal. Phytochemistry, 14, 2589-2592.

Heiler, C., Schieberle, P. (1997) Quantitative instrumental and sensory studies on aroma compounds contributing to a metallic flavour defect in buttermilk. International Dairy Journal, 7, 659-666.

Henning, H. (1916) Die qualitatenrihe des geschmacks. Z. Pscyhol., 203 - 219.

Heymann, H., Hedrick, H.B., Karrasch, M.A., Eggeman, M.K., Ellersieck, M.R. (1990) Sensory and chemical characteristics of fresh pork roasts cooked to different endpoint temperatures. J. Food Sci., 55 (3): 613-617.

Hymowitz, N., Eckholdt, H. (1996) Effects of a 2.5 mg silver acetate lozenge on initial and long term smoking cessation. Preventative Medicine, 24: 537-546.

International Labour Office (1983) Encyclopedia of occupational and health safety, Vol. I and II, 3rd ed.

ISO (1991) Sensory analysis. Methodology. Method of investigating sensitivity of taste. ISO Standard 3972.

Ivancevik, K.; Lunderquist, A.; Isaksson, A.; Hochberfs, P.; Wretlind, A. (1989) Clinical trials with a new iodinated lipid emulsion for computed topography of the liver. Acta Radiologica, 30(5): 449-57.

Jeremiah, L.E., Ball, R.O., Uttaro, B., Gibson, L.L. (1996) The relationship of chemical components to flavor attributes of bacon and ham. Food Research International, 29 (5/6): 4457-464.

Johnsen, P.B., Civille, G.V., Cervellotti, J.R. (1987) A lexicon of pond-raised catfish flavor descriptors. J. Sens. Stud., 2: 85-91.

Kachru, D.N., Tandon, S.K., Misra, U.K., Nag, D. (1989) Occupational lead poisoning among silver jewellery workers. Indian Journal of Medical Sciences, 43 (4):89-91.

Kim, T.H., Nguyen, T.T., Shin, J.H., Baek, H.H., Lee, H.J. (2000) Aroma-active compounds of miniature beefsteakplant (Mosla diantherea Maxim). J. Agric. Food Chem., 48: 2877-2881.

Kinnamon, S. C., Cummings, T. A. (1992) "Chemosensory transduction mechanisms in taste." Annual Review of Physiology, 54, 715-731.

Kumazawa, K., Masuda, H. (1999) Identification of potent odorants in Japanese green tea (Sen-cha). J. Agric. Food Chem, 47: 5169-5172.

Lindner, K., Czétényi, C., Kubát, K., Bolla, É., Szejtli, J. (1977) Sensory evaluation of a new dihyrochalcone sweetener: Part II - Studies in foods. Acta Alimnetaria, 6 (4): 323-330.

Linssen, J.P.H.; Janssens, J.L.G.M.; Roozen, J.P.; Posthumus, M.A. (1991) Sensory descriptors for a taint in water packed in test pouches made of polyethylene lined aluminum. Journal of Plastic Film and Sheeting, 7 (4): 294-305.

Lovell, R.T. (1983) Off-flavors in pond-cultured channel catfish. Wat. Scie. Tech, 15: 67-73.

Lundahl, D.S., Lukes, M. R., McDaniel, M. R., Henderson, L.A. (1986) A semiascending paired difference method for determining sensory threshold of added substances to background media. J. Sensory Studies, 1, 291-306.

Lyon, B.G. (1987) Development of chicken flavor descriptor attribute terms aided by multivariate statistical procedures. Journal of Sensory Studies, 2 (1): 55-67.

MacLeod, G., Coppock, B.M. (1978) Sensory properties of aroma of beef cooked conventionally and by microwave radiation. Journal of Food Science, 43: 145-161.

Mahadeviah, M.; Gowramma, R.V.; Naresh, R. (1983) Welded OTS cans as an alternative to soldered can. Journal of Food Science and Technology - India, 20 (5): 241-242.

Mahaffey, K.R., Corneliussen, P.E., Jelinek, C.F., Fiorino, J.A. (1975) Heavy metal exposure from foods. Environmental Health Perspectives, 12: 63-69.

Malcolm, R.; Currey, H.S.; Mitchell, M.A.; Keil, J.E. (1986) Sliver acetate gum as a deterrent to smoking. Chest, 90 (1): 107-11.

Mallevialle, J., Suffet, I.H. (1987) Identification and treatment of tastes and odors in drinking water. American Water Works Association, Denver, CO.

McBurney, D. H., Gent, J.F. (1979) "On the nature of taste qualities." Psychological Bulletin. 86 (1) 151-167.

Melchoir, W.R.; Jaber, L.A. (1996) Metformin: an antihyperglycemic agent for treatment of type II diabetes. The Annals of Pharmacotherapy, 30 (2): 158-164.

Morgan, B.H., Siu, R.G.H. (1957) Action of ionizing radiations on individual foods. The U.S. Army Research and Development Series No. 1, U.S. Army Quarter-master Corps, Washington.

Mozell, M. M., Smith, P. E., Sullivan, R. L. Jr., Swender, P. (1969) "A technique to occlude the nasal chemoreceptors during lingual flavor stimulation." Physiology and Behavior, 4, 131.

Mucklow, J.C., Bending, M.R., Kahn, G.C., Dollery, C.T. (1978) Drug concentration in saliva. Clin Pharmacol Ther. 24, 563.

Müller, R., Zschiesche, W.; Steffen, H.M.; Schaller, K.H. (1989) Tellurium - intoxication. Klin Wochenschr, 67: 1152-1155.

Nakagawa, M. (1972) Relationship between astringency and the reaction of astringents with proteins. Journal of the Japanese Society for Food Science and Technology, 19 (11): 531-537.

Norgen, R., Pfaffman, C. (1975) The pontine taste area in the rat. Brain Research. 91, 99-117.

Persson, T., von Sydow, E., Akesson, C. (1973) Aroma of canned beef: sensory properties. Journal of Food Science, 38: 386-392.

Physician's Desk Reference (1997) 51st ed. Montvale, NJ: Medical Economics Company.

Plattig, K. H. (1988) "The Sense of Taste" in Sensory Analysis of Food (Piggott, J.R., ed.). Elsevier Science Publishers, Ltd. London, Great Britain, pp. 1-24.

Poretta, S. (1995) Quality evaluation of canned whole tomatoes. Food Science and Technology International, 1: 97-104.

Portmann, M-O., Kilcast, D. (1996) Psychophysical characterization of new sweeteners of commercial importance for the EC food industry. Food Chemistry. 56 (3) 291-302.

Pritchard, T. C. (1991) "The primate gustatory systems" in Smell and Taste in Health and Disease (Getchell, T. V., Bartoshuk, L. M., Doty, R. L., and Snow, J.B. Jr., eds.), Raven Press, pp. 109-125.

Refsgaard, H.H.F., Brockoff, P.B., Jensen, B. (1998) Sensory and chemical changes in farmed Atlantic salmon. J. Agric. Food Chem., 46: 3473-3479.

Refsgaard, H.H.F., Brockoff, P.B., Jensen, B. (2000) Free polyunsaturated fatty acids cause taste deterioration of salmon during frozen storage. J. Agric. Food Chem, 48: 3280-3285.

Rice, W.H., McMahon, D.J. (1998) Chemical, physical, and sensory characteristics of Mozzarella cheese fortified using protein chelated iron or ferric chloride. J. Dairy Sci., 81: 318-326.

Risvik, E. (1986) Sensory evaluation of irradiated beef and bacon. J. of Sens. Studies, 1: 109-122.

Roper, S.D. (1989) The cell biology of vertebrate taste receptors. Ann. Rev. Neurosci. 12, 329-353.

Roy, G. (1992) A review of sweet taste potentiation brought about by divalent oxygen and sulfur incorporation. Crit. Rev. in Food Sci. and Nut., 31 (1/2): 59-77.

Rychlik, M., Schieberle, P., Grosch, W. (1998) Compilation of odor thresholds, odor qualities, and retention indices of key food odorants. Deutsche Forschungsanstalt feur Lebensmittelchemie: Garching, Germany.

Schiffman, S. S., Erickson, R. P. (1971) "A theoretical review: A psychophysical model for gustatory quality." Physiology and Behavior. 7, 617-633.

Schiffman, S. S.; Erickson, R. P. (1980) "The issue of primary tastes versus a taste continuum." Neuroscience and Biobehavioral Reviews, 4: 109-117.

Schiffman, S.S., Zervakis, J., Shaio, E., Heald, A.E. (1999) Effect of the nucleoside analogs zidovudine, didanosine, stavudine, and lamivudine on the sense of taste. Nutrition, 15 (11/12) pp. 854-859.

Shewfelt, R.L. (1981) Fish muscle lipolysis - a review. J. Food Biochem, 5: 79-100.

Shrader, J.H., Johnson, A.H. (1934) Freesing orange juice. Ind. Engng Chem, 26, 869-874.

Skramlick, E. von. (1926) Handbuch der physiologie der niederen sinne. Leipzig, Germany. Georg Threme, ed.

Smith-Swintosky, V. L., Plata-Salaman, C. R., Scott, T. R. (1991) "Gustatory neural coding in the monkey cortex: Stimulus quality." Journal of Neurophysiology. 66, 1156-1165.

Stein, R.B., Hanauer, S.B. (2000) Comparative tolerability of treatments for inflammatory bowel disease. Drug Safefy, 23 (5): 429-448.

Stephan, A., Steinhart, H. (1999) Identification of character impact odorants of different soybean lecithins. J. Agric. Food Chem, 47: 2854-2859.

Suffet, I.H.; Corado, A.; Chou, D.; McGuire, M.J.; Butterworth, S. (1996) AWWA taste and odor survey. Journal of American Water Works Association, 88 (4): 168-180.

Whitfield, F.B., Freeman, D.J., Last, K.H., Bannister, P.A., Kennett, B.H. (1982) Oct-1-en-3-ol and (5Z)-octa-1,5-dien-3-ol, compounds important in the flavour of prawns and sand-lobsters. Australian Journal of Chemistry, 35: 373-83.

Yanagisawa, K., Bartoshuk, L.M., Catalanotto, F.A., Lehman, C.D., Karrer, T.A., Kveton, J.F. (1998) Anestheisa of the chorda tympani nerve and taste phantoms. Physiology and Behavior, 63 (3) 329-335.

Yanagisawa, K., Bartoshuk, L.M., Karrer, T.A., Kveton, J.F., Catalanotto, F.A., Lehman, C.D., Weiffenbach, J.M. (1992) Anesthesia of the chorda tympani nerve: insights into a source of dysgeusia. Chemical Senses, 17, 724.

Yau, N.J.N., Huang, Y.J. (2000) The effect of membrane processed water on sensory properties of Oolong tea drinks. Food Quality and Preference, 11: 331-339.