5,8,11,14-Nonadecatetraenoic acid

• CAS: 3036-89-3
• Synonyms: 5,8,11,14-nonadecatetraenoic acid; 5,8,11,14-nonadecatetraenoic acid, (all Z)-isome
• Molecular Formula: C19H30O2
• Molecular Weight: 290.4 g/mol
• SMILES: CCCCC=CCC=CCC=CCC=CCCCC(=O)O
• InChI: InChI=1S/C19H30O2/c1-2-3-4-5-6-7-8-9-10-11-12-13-14-15-16-17-18-19(20)21/h5-6,8-9,11-12,14-15H,2-4,7,10,13,16-18H2,1H3,(H,20,21)/b6-5+,9-8+,12-11+,15-14+
• InChIKey: PPGFCKXUIAVTFL-SFYIKTPXSA-N

5,8,11,14-Nonadecatetraenoic acid (CAS 3036-89-3; synonym: 19:4 n-5, C19:4 ω5) is a C19 odd-chain polyunsaturated fatty acid possessing four methylene-interrupted cis double bonds at the Δ5,8,11,14 positions, identically matching the double-bond regiochemistry of arachidonic acid (20:4 n-6) but with a 19-carbon backbone. The compound occurs naturally in certain Mortierella fungi and marine organisms and has been employed as a mechanistic probe to dissect chain-length versus double-bond-position determinants in eicosanoid biosynthesis, cellular lipid incorporation, and hepatic lipid handling.

5,8,11,14-Nonadecatetraenoic Acid (CAS 3036-89-3) – Odd-Chain Arachidonic Acid Analog for Lipid Biochemistry and Eicosanoid Research Procurement

5,8,11,14-Nonadecatetraenoic acid (CAS 3036-89-3; synonym: 19:4 n-5, C19:4 ω5) is a C19 odd-chain polyunsaturated fatty acid possessing four methylene-interrupted cis double bonds at the Δ5,8,11,14 positions, identically matching the double-bond regiochemistry of arachidonic acid (20:4 n-6) but with a 19-carbon backbone [1]. The compound occurs naturally in certain Mortierella fungi and marine organisms [2] and has been employed as a mechanistic probe to dissect chain-length versus double-bond-position determinants in eicosanoid biosynthesis, cellular lipid incorporation, and hepatic lipid handling [3]. Its structural homology to arachidonic acid, combined with its odd-carbon chain length, creates quantifiable differentiation in enzyme substrate specificity and metabolic routing that cannot be replicated by even-chain C18 or C20 analogs, making targeted procurement essential for studies requiring discrimination between chain-length effects and double-bond-position effects.

• [1] Jakschik, B.A., et al. (1980) 'Fatty acid structural requirements for leukotriene biosynthesis', Prostaglandins, 20(2), pp. 401–410. doi: 10.1016/s0090-6980(80)80057-8. View Source
• [2] Shimizu, S., Kawashima, H., Akimoto, K., Shinmen, Y. and Yamada, H. (1991) 'Production of odd chain polyunsaturated fatty acids by Mortierella fungi', Journal of the American Oil Chemists' Society, 68(4), pp. 254–258. doi: 10.1007/BF02657620. View Source
• [3] Ikeda, I., Murakami, J., Oka, T., Sugano, M., Yamada, H., Shimizu, S., Kawashima, H., Shinmen, Y. and Amachi, T. (1991) 'Effects of oleic, arachidonic and 5,8,11,14-nonadecatetraenoic acids on lipid secretion and ketogenesis in perfused rat liver', Lipids, 26(1), pp. 27–30. doi: 10.1007/BF02544020. View Source

Why 5,8,11,14-Nonadecatetraenoic Acid Cannot Be Replaced by Arachidonic Acid or Other Even-Chain PUFA Analogs in Mechanistic Studies

Generic substitution with arachidonic acid or other even-chain polyunsaturated fatty acids is demonstrably inadequate for studies targeting the chain-length specificity of phospholipase A₂-mediated deacylation, eicosanoid biosynthesis, or intracellular lipid trafficking, because agonist-sensitive phospholipases release 5,8,11,14-tetraenoic fatty acids with 18, 19, 20, and 21 carbons without stringent chain-length discrimination [1], yet the metabolic fate of the released fatty acid diverges sharply: 19:4 n-5 is preferentially elongated to C21 and C23 PUFAs in hepatic cells rather than being processed exclusively through eicosanoid pathways [2], and it competes directly with 20:4 n-6 for incorporation into keratinocyte phospholipids more effectively than any other tested analog [3]. This dual property—chain-length-permissive release but chain-length-discriminative downstream metabolism—creates a scientific requirement for the authentic C19 compound that no even-chain surrogate can fulfill.

• [1] Rosenthal, M.D. and Hill, J.R. (1989) 'Substrate specificity of the agonist-stimulated release of polyunsaturated fatty acids from vascular endothelial cells', Archives of Biochemistry and Biophysics, 271(2), pp. 346–355. doi: 10.1016/0003-9861(89)90474-8. View Source
• [2] Sakuradani, E., Kobayashi, M. and Shimizu, S. (2000) 'C19 Odd-chain Polyunsaturated Fatty Acids (PUFAs) Are Metabolized to C21-PUFAs in a Rat Liver Cell Line, and Curcumin, Gallic Acid, and Their Related Compounds Inhibit Their Desaturation', Bioscience, Biotechnology, and Biochemistry, 64(8), pp. 1641–1650. doi: 10.1271/bbb.64.1641. View Source
• [3] Nakano, N., Shirasaka, N., Murakami, T., Shimizu, S. and Yoshizumi, H. (2001) 'Competitive incorporation of arachidonic acid analogs by cultured rat keratinocytes', Journal of Nutritional Science and Vitaminology, 47(1), pp. 69–77. doi: 10.3177/jnsv.47.69. View Source

5,8,11,14-Nonadecatetraenoic Acid – Comparator-Anchored Quantitative Differentiation Evidence for Compound Selection

Hepatic Triglyceride Secretion and Ketone Body Production: 19:4 n-5 Retains Oleic Acid-Like Metabolic Profile, Distinct from Arachidonic Acid

In isolated perfused rat liver, 5,8,11,14-nonadecatetraenoic acid (19:4 n-5, 1.2 mM perfusate) maintained ketone body production and triglyceride (TG) secretion at levels comparable to oleic acid (18:1 n-9), while arachidonic acid (20:4 n-6) at the same concentration significantly reduced both parameters [1]. Post-perfusion hepatic TG concentration was also lower with arachidonic acid than with either oleic acid or 19:4 n-5, and cholesterol secreted into the perfusate was significantly higher with both tetraenoic acids (19:4 n-5 and 20:4 n-6) compared to the monounsaturated oleic acid control [1]. Critically, the concentration of cholesterol retained in the post-perfused liver was slightly but significantly higher with 19:4 n-5 than with either oleic acid or arachidonic acid (20:4 n-6), indicating a unique hepatic cholesterol-handling property specific to the C19 tetraenoic acid [1].

Evidence Dimension	Perfusate-stimulated hepatic triglyceride secretion, ketone body production, and post-perfusion liver cholesterol retention
Target Compound Data	19:4 n-5 (1.2 mM): TG secretion and ketone body production comparable to oleic acid; cholesterol secretion significantly elevated vs. oleic acid; post-perfusion hepatic cholesterol concentration significantly higher than both oleic acid and arachidonic acid
Comparator Or Baseline	Arachidonic acid (20:4 n-6, 1.2 mM): TG secretion and ketone body production significantly lower than oleic acid and 19:4 n-5 (p<0.05); Oleic acid (18:1 n-9, 1.2 mM): TG secretion and ketone body production significantly higher than arachidonic acid, comparable to 19:4 n-5
Quantified Difference	Ketone body production and TG secretion: 19:4 n-5 ≈ oleic acid >> arachidonic acid (statistically significant). Post-perfusion liver cholesterol: 19:4 n-5 > oleic acid ≈ arachidonic acid (p<0.05).
Conditions	Isolated perfused rat liver (male Sprague-Dawley rats), 1.2 mM fatty acid–albumin complex in Krebs-Henseleit bicarbonate buffer, 60 min recirculating perfusion, n=5 per group

Why This Matters

For researchers modeling fatty acid effects on hepatic VLDL assembly and ketogenesis, substituting arachidonic acid for 19:4 n-5 would introduce a confounding suppression of TG secretion and ketone body production, whereas 19:4 n-5 uniquely captures the tetraenoic effect on cholesterol secretion without the TG-suppressive effect, enabling clean dissection of double-bond-number effects from chain-length effects on hepatic lipid handling.

• [1] Ikeda, I., Murakami, J., Oka, T., Sugano, M., Yamada, H., Shimizu, S., Kawashima, H., Shinmen, Y. and Amachi, T. (1991) 'Effects of oleic, arachidonic and 5,8,11,14-nonadecatetraenoic acids on lipid secretion and ketogenesis in perfused rat liver', Lipids, 26(1), pp. 27–30. doi: 10.1007/BF02544020. View Source

Competitive Incorporation into Keratinocyte Phospholipids: 19:4 n-5 Is the Most Effective Competitor Against Arachidonic Acid Incorporation Among Six Tested PUFAs

In a fetal rat skin keratinocyte (FRSK) cell line, 5,8,11,14-nonadecatetraenoic acid (19:4 n-5) was compared head-to-head with five other polyunsaturated fatty acids—mead acid (20:3 n-9), dihomo-γ-linolenic acid (20:3 n-6), 11,14,17-eicosatrienoic acid (20:3 n-3), arachidonic acid (20:4 n-6), and eicosapentaenoic acid (20:5 n-3)—for competitive incorporation into total cellular lipids and phospholipids over a 24 h incubation [1]. When 19:4 n-5 and 20:4 n-6 were co-incubated at a 0.5:0.5 ratio, 19:4 n-5 reduced the incorporation of 20:4 n-6 into phospholipids more effectively than any other analog tested [1]. In single-substrate incubations, 19:4 n-5 was preferentially incorporated into total cellular lipids and phospholipids to a degree comparable to 20:4 n-6 itself, and significantly greater than 20:3 n-9 or 20:5 n-3 [1].

Evidence Dimension	Competitive reduction of 20:4 n-6 incorporation into keratinocyte phospholipids under equimolar co-incubation
Target Compound Data	19:4 n-5 (at 0.5:0.5 ratio with 20:4 n-6): most effective inhibitor of 20:4 n-6 incorporation into phospholipids among all six tested PUFAs
Comparator Or Baseline	20:3 n-9, 20:3 n-6, 20:3 n-3, 20:5 n-3 (each at 0.5:0.5 ratio with 20:4 n-6): all less effective than 19:4 n-5 at reducing 20:4 n-6 phospholipid incorporation
Quantified Difference	19:4 n-5 > 20:5 n-3 ≈ 20:3 n-9 for competitive displacement of 20:4 n-6 from phospholipids (rank order established by capillary gas-liquid chromatography; exact percentage reduction unpublished outside full text)
Conditions	Fetal rat skin keratinocyte (FRSK) cell line, 24 h incubation, PUFA ratio 1:0, 0.5:0.5, or 0:1, lipid extraction and phospholipid analysis by capillary GLC

Why This Matters

For investigators studying epidermal lipid biology or atopic dermatitis models where arachidonic acid incorporation into keratinocyte phospholipids drives prostaglandin and leukotriene production, 19:4 n-5 provides a unique tool—it competes for the same phospholipid incorporation machinery more effectively than Mead acid or EPA, yet its odd-chain backbone generates non-canonical prostaglandin homologs, allowing pharmacological uncoupling of membrane incorporation from classical eicosanoid signaling.

• [1] Nakano, N., Shirasaka, N., Murakami, T., Shimizu, S. and Yoshizumi, H. (2001) 'Competitive incorporation of arachidonic acid analogs by cultured rat keratinocytes', Journal of Nutritional Science and Vitaminology, 47(1), pp. 69–77. doi: 10.3177/jnsv.47.69. View Source

Leukotriene Biosynthesis Substrate Recognition: 19:4 (5,8,11,14) Is Readily Converted by 5-Lipoxygenase, Matching Arachidonic Acid and Outperforming Δ4, Δ6, Δ7, or Δ8 Fatty Acids

Using a cell-free enzyme system from RBL-1 cells, Jakschik et al. (1980) systematically evaluated the structural requirements for leukotriene biosynthesis across fatty acids varying in chain length and unsaturation position [1]. 5,8,11,14-Nonadecatetraenoic acid (19:4) was converted to products comigrating with 5-HETE and 5,12-DiHETE and generated biologically active slow-reacting substances (SRS, i.e., leukotrienes) to an extent comparable to arachidonic acid (20:4) and eicosapentaenoic acid (20:5), which share the same Δ5,8,11 unsaturation pattern [1]. In contrast, fatty acids with the initial unsaturation at Δ4, Δ6, Δ7, or Δ8 were poor substrates for the leukotriene enzyme system [1]. Chain length per se (19 vs. 20 carbons) did not influence conversion, demonstrating that the Δ5,8,11 motif is the dominant recognition determinant [1].

Evidence Dimension	Conversion to 5-HETE/5,12-DiHETE comigrating products and biologically active SRS (leukotrienes) by leukocyte 5-lipoxygenase
Target Compound Data	19:4 (5,8,11,14-nonadecatetraenoic acid): readily converted to 5-HETE/5,12-DiHETE-type products and biologically active SRS, comparable to 20:4
Comparator Or Baseline	20:4 (arachidonic acid), 20:5 (eicosapentaenoic acid), 20:3 (5,8,11-eicosatrienoic acid): all readily converted; 18:4 (5,8,11,14-octadecatetraenoic acid): also converted; Fatty acids with Δ4, Δ6, Δ7, or Δ8 initial unsaturation: poor substrates
Quantified Difference	Δ5,8,11 motif-containing fatty acids (including 19:4): active substrates (qualitatively comparable product formation); Non-Δ5,8,11 fatty acids: poor substrates (product formation qualitatively negligible)
Conditions	Cell-free enzyme system from RBL-1 rat basophilic leukemia cells, radiolabeled substrate, product analysis by HPLC comigration with authentic 5-HETE and 5,12-DiHETE standards, biological SRS activity assay on guinea pig ileum

Why This Matters

For researchers interrogating 5-lipoxygenase substrate specificity, 19:4 n-5 serves as the only commercially accessible odd-chain tetraenoic fatty acid that is an authentic substrate for leukotriene biosynthesis, enabling experimental designs that distinguish between the contribution of the Δ5,8,11 pharmacophore and the C20 carbon skeleton to leukotriene-dependent inflammatory signaling, without the confounding variable of altered double-bond regiochemistry that accompanies substitution with Mead acid or EPA.

• [1] Jakschik, B.A., Sams, A.R., Sprecher, H. and Needleman, P. (1980) 'Fatty acid structural requirements for leukotriene biosynthesis', Prostaglandins, 20(2), pp. 401–410. doi: 10.1016/s0090-6980(80)80057-8. View Source

Prostaglandin Conversion Efficiency Ranking: 19:4 ω5 Occupies a Quantitatively Defined Intermediate Position Between 20:4 ω6 and 19:3 ω5 in Sheep Vesicular Gland Cyclooxygenase

Van Dorp and colleagues (1968) synthesized a panel of odd-numbered and even-numbered polyenoic fatty acids and quantitatively compared their enzymatic conversion to prostaglandin homologs by a particulate fraction of sheep vesicular glands [1]. The conversion rates, measured as yield of biologically active prostaglandin-like material, followed the rank order: 20:4 ω6 (arachidonic acid) > 20:3 ω6 (dihomo-γ-linolenic acid) > 21:4 ω7 > 19:4 ω5 > 21:3 ω7 > 19:3 ω5 [1]. 5,8,11,14-Nonadecatetraenoic acid (19:4 ω5) yielded ω-homo PGE₁, a homolog of prostaglandin E₁ with biological activity on guinea pig ileum, rat blood pressure, and ADP-induced platelet aggregation comparable to native PGE₁ [1]. The essential fatty acid (EFA) activity in the rat correlated roughly with the prostaglandin conversion rate [1].

Evidence Dimension	Relative enzymatic conversion rate to biologically active prostaglandin homologs by sheep vesicular gland particulate fraction
Target Compound Data	19:4 ω5: rank 4 of 6 fatty acids tested; yields ω-homo PGE₁; EFA activity positive (considerable)
Comparator Or Baseline	20:4 ω6 (arachidonic acid): rank 1 (highest conversion rate); 20:3 ω6 (dihomo-γ-linolenic acid): rank 2; 21:4 ω7: rank 3; 21:3 ω7: rank 5; 19:3 ω5: rank 6 (lowest conversion rate); 18:3 ω4 and 22:3 ω8: no conversion detected
Quantified Difference	Conversion rate rank order: 20:4 ω6 > 20:3 ω6 > 21:4 ω7 > 19:4 ω5 > 21:3 ω7 > 19:3 ω5 (exact percentage conversion values not reported in abstract; relative ranking based on biological activity of prostaglandin products)
Conditions	Particulate fraction of sheep vesicular glands (cyclooxygenase source), fatty acid substrate incubation, prostaglandin product detection by guinea pig ileum contraction, rat blood pressure assay, and ADP-induced platelet aggregation

Why This Matters

For researchers building quantitative structure-activity relationships (QSAR) for cyclooxygenase substrates, 19:4 ω5 occupies a precisely defined intermediate position that fills the gap between the high-conversion C20 substrates (20:4, 20:3) and the low-conversion C19 trienoic acid (19:3), providing a unique data point for modeling the contribution of the fourth double bond and chain length to cyclooxygenase catalytic efficiency.

• [1] van Dorp, D.A., van der Vies, J., Houtsmuller, U.M.T. and van der Beek, J.R. (1968) 'Biologically active prostaglandins from some new odd-numbered essential fatty acids', Recueil des Travaux Chimiques des Pays-Bas, 87(4), pp. 405–416. doi: 10.1002/recl.19680870414. View Source

Microbial Production Benchmark: Mortierella alpina 1S-4 Accumulates 5,8,11,14-Nonadecatetraenoic Acid at 44.4 mg/g Dry Mycelia (11.2% of Total Fatty Acids) Under Defined Fermentation Conditions

Shimizu et al. (1991) screened multiple arachidonic acid-producing Mortierella strains and identified Mortierella alpina 1S-4 as a high producer of odd-chain PUFAs [1]. With 5% (w/v) n-heptadecane as the odd-chain carbon precursor and 1% yeast extract as the nitrogen source, the fungus accumulated 5,8,11,14-nonadecatetraenoic acid (C19:4) at a titer of 44.4 mg/g dry mycelia (0.68 mg/mL culture broth), constituting 11.2% of total extracted mycelial fatty acids, with odd-chain fatty acids comprising over 95% of the total [1]. Addition of sesamin, a specific Δ5 desaturase inhibitor, increased C19:3 and decreased C19:4, confirming the biosynthetic route: C17:0 → C17:1 → C17:2 → C17:3 → C19:3 → C19:4, which mimics the n-6 pathway to arachidonic acid [1].

Evidence Dimension	Microbial production titer of odd-chain C19:4 PUFA under optimized fermentation
Target Compound Data	C19:4 (5,8,11,14-nonadecatetraenoic acid): 44.4 mg/g dry mycelia, 0.68 mg/mL culture broth, 11.2% of total fatty acids, odd-chain fatty acids >95% of mycelial total
Comparator Or Baseline	Other Mortierella strains that produce C20 PUFAs (arachidonic acid): unable to accumulate C19 PUFAs without odd-chain substrate feeding; species lacking C20 PUFA biosynthetic capacity: accumulated C17 but not C19 PUFAs under odd-chain substrate conditions
Quantified Difference	C19:4 titer of 44.4 mg/g under precursor-fed conditions establishes a documented procurement-sourcing benchmark absent from synthetic chemical suppliers
Conditions	Mortierella alpina 1S-4, 5% n-heptadecane + 1% yeast extract, submerged fermentation (culture broth volume unspecified in abstract), fatty acid extraction and analysis by GC

Why This Matters

For laboratories requiring multi-gram quantities of 5,8,11,14-nonadecatetraenoic acid, this fermentation benchmark provides a documented bioproduction reference point for evaluating custom fermentation sourcing versus synthetic procurement, and the odd-chain enrichment strategy (>95% odd-chain purity in total mycelial lipids) offers a purification advantage over chemical synthesis that typically yields complex isomer mixtures.

• [1] Shimizu, S., Kawashima, H., Akimoto, K., Shinmen, Y. and Yamada, H. (1991) 'Production of odd chain polyunsaturated fatty acids by Mortierella fungi', Journal of the American Oil Chemists' Society, 68(4), pp. 254–258. doi: 10.1007/BF02657620. View Source

Procurement-Driven Application Scenarios for 5,8,11,14-Nonadecatetraenoic Acid Based on Comparator-Anchored Quantitative Evidence

Dissecting Chain-Length vs. Double-Bond Contributions to Hepatic VLDL Assembly and Ketogenesis

In isolated perfused rat liver, arachidonic acid (20:4 n-6) suppresses both triglyceride secretion and ketone body production relative to oleic acid, while 19:4 n-5 does not, despite both bearing the identical Δ5,8,11,14 tetraene system [1]. This makes 19:4 n-5 uniquely suited for experiments requiring a tetraenoic fatty acid that permits normal rates of hepatic VLDL-TG secretion and ketogenesis, enabling investigators to attribute the suppressive effect of arachidonic acid specifically to its 20-carbon chain length or its downstream eicosanoid products rather than to tetraenoic unsaturation per se.

Pharmacological Perturbation of Arachidonic Acid Incorporation into Keratinocyte Phospholipids in Skin Inflammation Models

In fetal rat skin keratinocytes, 19:4 n-5 was the most effective competitor against arachidonic acid for incorporation into total cellular and phospholipid pools among all six PUFAs tested, including Mead acid and EPA [2]. This competitive property, combined with its generation of non-canonical prostaglandin homologs rather than PGE₂ or PGD₂ [3], positions 19:4 n-5 as a tool compound for selectively depleting keratinocyte phospholipid arachidonate pools without recapitulating the full inflammatory eicosanoid cascade, offering a more targeted probe than EPA or DHA for atopic dermatitis and psoriasis mechanistic studies.

Validating 5-Lipoxygenase Substrate Specificity Models Independent of Chain Length

The RBL-1 cell-free leukotriene biosynthesis system converts 19:4 (5,8,11,14) to 5-HETE, 5,12-DiHETE, and biologically active SRS with efficiency comparable to 20:4, while fatty acids lacking the Δ5,8,11 motif fail as substrates [4]. 19:4 n-5 thus serves as the minimal tetraenoic probe that retains full 5-lipoxygenase substrate competency, enabling studies that require a radiolabeled or chemically modified odd-chain tetraenoic acid to track leukotriene metabolite fate by mass spectrometry without the mass overlap that complicates even-chain arachidonic acid tracer experiments.

Building Quantitative Cyclooxygenase Substrate Selectivity Models Across C19–C21 Odd-Chain Fatty Acids

The precisely defined rank order of prostaglandin conversion rates—20:4 ω6 > 20:3 ω6 > 21:4 ω7 > 19:4 ω5 > 21:3 ω7 > 19:3 ω5—established by van Dorp et al. (1968) positions 19:4 ω5 as a bridge data point between high-activity C20 substrates and lower-activity C19/C21 trienoic acids [3]. Procurement of 19:4 ω5 alongside the comparator fatty acids in this series enables direct experimental replication of this historical rank order under modern LC-MS/MS analytical conditions, facilitating updated quantitative kinetic parameter determination (Km, Vmax) for a structure-activity landscape that has remained largely unexplored since 1968.

• [1] Ikeda, I., Murakami, J., Oka, T., Sugano, M., Yamada, H., Shimizu, S., Kawashima, H., Shinmen, Y. and Amachi, T. (1991) 'Effects of oleic, arachidonic and 5,8,11,14-nonadecatetraenoic acids on lipid secretion and ketogenesis in perfused rat liver', Lipids, 26(1), pp. 27–30. doi: 10.1007/BF02544020. View Source
• [2] Nakano, N., Shirasaka, N., Murakami, T., Shimizu, S. and Yoshizumi, H. (2001) 'Competitive incorporation of arachidonic acid analogs by cultured rat keratinocytes', Journal of Nutritional Science and Vitaminology, 47(1), pp. 69–77. doi: 10.3177/jnsv.47.69. View Source
• [3] van Dorp, D.A., van der Vies, J., Houtsmuller, U.M.T. and van der Beek, J.R. (1968) 'Biologically active prostaglandins from some new odd-numbered essential fatty acids', Recueil des Travaux Chimiques des Pays-Bas, 87(4), pp. 405–416. doi: 10.1002/recl.19680870414. View Source
• [4] Jakschik, B.A., Sams, A.R., Sprecher, H. and Needleman, P. (1980) 'Fatty acid structural requirements for leukotriene biosynthesis', Prostaglandins, 20(2), pp. 401–410. doi: 10.1016/s0090-6980(80)80057-8. View Source