Citrus bitterness is not entirely decided the instant the fruit is squeezed. Within the same glass of juice, something that is already bitter at the moment of pressing coexists with something that turns bitter only some time after pressing. The former is called immediate bitterness, carried by flavanone glycosides such as naringin; the latter is called delayed bitterness, carried by limonoids such as limonin. This distinction is not just a matter of classification for its own sake. Which one shows up more strongly changes by cultivar, and the debittering process itself becomes a different thing depending on which one it targets.
It has been confirmed that citrus bitterness comes from two kinds with different origins: an "immediate" bitterness from compounds such as naringin that take effect right after squeezing, and a "delayed" bitterness from limonin that thickens only some time after squeezing.
It has been confirmed, by direct measurement of orange juice stored after harvest, that limonin exists before pressing as a non-bitter precursor (LARL) and converts, after pressing, into bitter limonin.
Classic sensory tests have measured that limonin's bitterness is hardest to perceive (its threshold is highest) at pH 3.8, and that bitterness becomes more noticeable on either side of this optimum.
Whether the conversion into limonin is a chemical reaction driven by acid or one driven by an enzyme (LDLH) is explained differently between studies, and has not yet been settled one way.
1. Bitterness Is Not One Thing
Taste-reconstitution experiments on orange juice could not find a single adequate marker for bitterness. Limonin, limonoid glycosides, polymethoxyflavones, organic acids, and sugars intertwine to create the bitterness.
orange juice — bitterness is generated by: The interplay of limonin, limonoid glycosides, polymethoxyflavones, organic acids, and sugarssupported
Scope: Activity-guided fractionation of orange juice and taste reconstitution
Note: A single compound was judged insufficient as a bitterness marker.
Taste re-engineering experiments showed a complex interplay, and no single adequate marker for the bitterness of orange juice was found.
Limitations: The exact juice cultivar and panel details require checking the full text.
A panel of 20-27 screened judges determined the thresholds of limonin and naringin in water, sucrose solution, citric acid solution, and a citrus-juice model system. The threshold in distilled water was 1 mg/l for limonin and 20 mg/l for naringin, and sucrose, citric acid, and their combination raised both thresholds several-fold. In a model with 1.2% citric acid plus 5% sucrose, 2.5% dextrose, and 2.5% fructose, limonin's maximum threshold of 11 mg/kg occurred at pH 3.8, and the maximum recognition threshold for limonin in natural orange juice was 6.5 mg/kg, also at pH 3.8. No pH optimum was seen for naringin.
Limitations: Supports only the acid and sugar side of the interplay. Polymethoxyflavones and limonoid glycosides were not examined. The abstract was reconstructed from OpenAlex (single pathway). 1973 methodology.
Context: Verbatim: 'There are two types of bitterness, namely immediate bitterness and delayed bitterness, in citrus fruits, imparted by two different types of compounds.'; 'The immediate bitterness is largely conferred by naringin and neohesperidin.'; 'the delayed bitterness is mainly produced by limonin of limonoids.'; 'The bitter-tasting naringin is the major contributor to the bitter taste in grapefruit and neohesperidin in sour orange.' In this way, an independent review likewise attributes the bitterness of citrus juice not to a single marker compound but to multiple compound classes (flavanone neohesperidosides and limonoids).
Limitations: The flavanone contributors named belong to bitter species (pomelo, grapefruit, bitter/sour orange) and are largely absent from sweet orange. Because polymethoxyflavones are not singled out as bitterness contributors, this review does not support the PMF portion of the interplay finding above.
Context: As a sentence absent from the PubMed abstract, states that the effect of sugars on the perceived bitterness of limonoids and PMFs in orange juice was "quantitatively elucidated." This pairs with the existing finding of sub-threshold bitterness enhancement by PMFs, suggesting an as-yet-unrecorded direction — bitterness suppression or modulation by sugars. Verbatim: "For the first time, sub-threshold concentrations of PMFs were shown to enhance the perceived bitterness of limonoids. Moreover, the influence of sugars on the perceived bitterness of limonoids and PMFs in orange juice relevant concentration ranges was quantitatively elucidated."
Limitations: Because the full text has not been obtained, whether the sugar effect runs toward suppression or enhancement, the quantitative values, and the sugars involved (sucrose, fructose, glucose, etc.) are unknown. This is only a lead found from a difference in how the abstract was paraphrased; turning it into a new claim would need the full text checked.
Show dataingredient-orange-juice —has_bitterness_generated_by→ "The interplay of limonin, limonoid glycosides, polymethoxyflavones, organic acids, and sugars"
Within that interplay, naringin is counted as one of citrus's two major bitter compounds, and has been treated as a component that contributes to immediate bitterness.
Limits: Naringin and limonin have been "long known to play a key role in the bitterness of grapefruit," yet naringin does not activate TAS2R19 in vitro, and the receptor implicated by grapefruit-liking genetics (TAS2R31) has been verified with quinine, not naringin; the receptor-level basis of naringin perception remains unresolved in this line of literature.
Limitations: Does not dispute that naringin itself is bitter; it says grapefruit bitterness cannot be assigned at the receptor level to any single characterized TAS2R.
Limonin, by contrast, exists at the moment of pressing as a non-bitter precursor, and converts over time into a bitter limonoid. This delay is what the term "delayed bitterness" refers to.
Scope: In many of the citrus products where it occurs, a delayed bitterness contribution
Note: Matrix and precursor conversion affect how it is perceived over time.
Verbatim: "The gradual conversion of the non-bitter precursors to the bitter limonoids is known as the process of delayed bitterness." During juicing, "the precursors NARL and LARL are converted to the bitter nomilin and limonin, respectively, under the acidic condition (pH < 6.5)", and this is accelerated once membrane vesicles rupture, by limonoid D-ring lactone hydrolase. Citrus limonoids exist as "the bitter aglycones and non-bitter glucosides", and limonoid aglycones are partly converted to glucosides during fruit maturation (a natural debittering by limonoid glucosyltransferase).
Limitations: A review; gives no numerical thresholds. The review names naringin (grapefruit) and neohesperidin (sour orange) as the main flavanone bitterness contributors, and limonin as the delayed-bitterness contributor.
Limonin's bitterness threshold in distilled water was 1 mg/l and rose several-fold in the presence of sucrose and citric acid; the optimum pH at which limonin's bitterness suppression is greatest (the threshold is highest) was 3.8 both in the model system and in natural orange juice (maximum recognition threshold 6.5 mg/kg), and on either side of that pH the threshold fell, making the bitterness more noticeable.
Limitations: Quantifies the matrix-dependence of limonin's bitterness but does not address the kinetics of delayed precursor conversion. The abstract was reconstructed from OpenAlex (single pathway).
Delayed bitterness is not a figure of speech; it is used as a term for a chemical conversion process. In the juice after pressing, the non-bitter precursor limonoate A-ring lactone (LARL) converts into bitter limonin.
limonin is produced through: In citrus juice after pressing, the conversion of the non-bitter precursor limonoate A-ring lactone (LARL) into bitter limoninsupported
Scope: Citrus juice that develops delayed bitterness during storage after pressing
The limonin content of freshly squeezed navel orange juice rose during storage and reached a constant maximum value; the paper describes delayed bitterness as limonin formation, after pressing, from the non-bitter precursor LARL.
Limitations: A single seedless cultivar; no human sensory measurement.
States as background that citrus's delayed bitterness arises from the conversion of the non-bitter precursor limonoate A-ring lactone (LARL) into limonin.
Limitations: Abstract only. This statement frames the enzyme study; it is not a new demonstration.
Show datacompound-limonin —is_produced_through→ "In citrus juice after pressing, the conversion of the non-bitter precursor limonoate A-ring lactone (LARL) into bitter limonin"
However, what drives this conversion is split between sources. A study on navel orange juice assumes it to be an acid-promoted reaction, while another study frames it as catalysis by an enzyme called limonin D-ring lactone hydrolase. Neither has directly settled the question within the juice itself.
limonin — currently proposed mechanisms: Acid-promoted lactonization of LARL versus conversion catalyzed by limonin D-ring lactone hydrolase (LDLH)contested
Scope: Post-pressing limonin formation in citrus juice. Kinetic evidence from seedless Newhall navel orange juice versus an enzyme purified from seeded sweet-orange seeds
Note: Zhang 2018 concludes the reaction is acid-promoted and likely excludes enzyme involvement in seedless juice. Zhang 2024 places LDLH catalysis as the source of delayed bitterness formation. Whether seeds are present could reconcile the two, but no source read here states this explicitly.
The authors take limonin formation from LARL in navel orange juice as one that "should be an acid-promoted reaction which was accelerated by the addition of acid and heat treatment", and likely exclude the involvement of limonoid D-ring lactone hydrolase. The authors note that this enzyme's activity has been reported only in citrus seeds, and that Newhall fruit is normally seedless.
Limitations: The exclusion of enzyme involvement is an inference from reaction kinetics and from reports of the enzyme's localization in seeds, not a direct enzyme assay in the juice itself.
Contradicts: Describes delayed bitterness as arising from the conversion of LARL to limonin "under the catalysis of limonin D-ring lactone hydrolase (LDLH)", and purifies a pH- and heat-stable LDLH (active pH 3-9, stable for 120 minutes at 50°C) from sweet-orange seeds.
Limitations: Abstract only; since the enzyme comes from seeds, this discrepancy may not apply to seedless juice. Neither paper read here resolves the point.
Show datacompound-limonin —has_currently_proposed_mechanisms→ "Acid-promoted lactonization of LARL versus conversion catalyzed by limonin D-ring lactone hydrolase (LDLH)"
A study that tracked navel orange juice also recorded that the maximum value of limonin itself falls as the post-harvest storage period lengthens. Heat processing and acidification only sped up the formation; they did not change the maximum value reached.
limonin — juice content is affected by: Post-harvest fruit storage time (the juice maximum fell from about 24.24 to 1.04 mg/L over 0-45 days), heat processing, pH, and filtrationsupported
Scope: Newhall navel orange juice stored after pressing. 3 harvest batches
Note: Heat processing and acidification only sped up limonin formation; they did not change the maximum steady-state value itself. Filtration and longer fruit storage lowered the maximum value.
Limonin's maximum constant value (MCV) fell with time after harvest (about 24.24, 6.68, 4.15, and 1.04 mg/L for fruit pressed 0, 9, 25, and 45 days after harvest). Heat processing and acidification only sped up limonin formation and did not change the MCV. Filtered juice had lower limonin (1.15 versus 1.30 mg/L).
Limitations: A single cultivar from a single origin. The mg/L values are chemical content, not linked to a sensory panel.
Show datacompound-limonin —has_juice_content_affected_by→ "Post-harvest fruit storage time (the juice maximum fell from about 24.24 to 1.04 mg/L over 0-45 days), heat processing, pH, and filtration"
The concentration at which bitterness is perceived also differs in level between limonin and naringin. The threshold in distilled water is lower for limonin, while naringin is perceived as bitter only at a much higher concentration.
limonin — bitter recognition threshold: 1 mg/L in distilled water. Maximum recognition threshold of 6.5 mg/kg in orange juice at pH 3.8supported
Scope: A panel of 20-27 screened judges. Distilled water, sucrose, and citric acid model systems, and orange juice
Note: Sucrose and citric acid raise the threshold. In both the model systems and the juice, the pH optimum for suppression is 3.8. Later debittering literature cites the 6.5 ppm juice value from this same lineage.
The threshold in distilled water was 1 mg/l for limonin and 20 mg/l for naringin. The maximum recognition threshold for limonin in orange juice was 6.5 mg/kg, occurring at pH 3.8, and the maximum value of 11 mg/kg in the model system also occurred at pH 3.8.
Limitations: Abstract only; panel procedure details and inter-individual variation cannot be read.
Show datacompound-limonin —has_bitter_recognition_threshold→ "1 mg/L in distilled water. Maximum recognition threshold of 6.5 mg/kg in orange juice at pH 3.8"
naringin — bitter recognition threshold: 20 mg/L in distilled water. In juice, it is stated that immediate bitterness arises once naringin exceeds about 20 ppmsupported
Scope: The value in water comes from a panel of 20-27 screened judges. The statement about juice derives from debittering literature citing the same lineage
Naringin's threshold in distilled water was 20 mg/l. No clear pH optimum for naringin bitterness suppression was reported (limonin's optimum is pH 3.8).
Limitations: Abstract only. The naringin threshold in a juice matrix is not given in the abstract that could be read.
Context: States that naringin is extracted into the juice from the membranes and the albedo, and that once its concentration exceeds 20 ppm it gives an immediate bitterness.
Limitations: A secondhand citation (the source text has the misspelling 'narinigin'). Not an independent sensory measurement.
Show datacompound-naringin —has_bitter_recognition_threshold→ "20 mg/L in distilled water. In juice, it is stated that immediate bitterness arises once naringin exceeds about 20 ppm"
A study analyzing hassaku juice and peel detected limonoid glycosides, mainly limonin 17-β-D-glucopyranoside. Content was especially high in the dried inner and outer peel.
Scope: Hassaku fruit (juice, inner peel, outer peel); analysis by TLC and HPLC
Note: Note that these are not the bitter limonin (aglycone) itself, but glycosides whose bitterness has been weakened through glucose conjugation. This connects, as data on debittering through sugar conjugation during ripening, separately from the existing claim (compound-limonin's delayed bitterness).
Separated and quantified 6 limonoid glycosides in the juice, inner peel, and outer peel of hassaku fruit by TLC and HPLC, and reported that limonin 17-β-D-glucopyranoside accounted for 148 ppm of the 314 ppm total in the juice (47%). 3100 ppm in the dried inner peel, 2300 ppm in the dried outer peel.
Limitations: What was detected were the weakly bitter glycosides (the glucose-conjugated form); a direct correspondence with bitterness sensory perception could not be confirmed from this abstract alone.
Source: ハッサク果実のリモノイド配糖体 (前田久夫; 尾崎嘉彦; 三宅正起; 伊福靖; 長谷川信, 1990) · full text checked
Show dataingredient-hassaku —contains→ compound-limonin
Scope: Direct measurement of natsumikan first-press juice (Nihon Kankitsu Kogyo K.K., Yawatahama factory, batch produced in mid-March)
Note: A 1962 analysis. Note that the content differs by more than twofold between pressing stages (1st vs. 2nd press).
Measured naringin at 75.2 mg% in the 1st-press juice of natsumikan (glucose 2.6%, sucrose 2.6%, citric acid 2.9%, pH 2.9), and 156 mg% in the 2nd-press juice (sugar 5.1%, pH 2.9).
Limitations: Information on the growing region and cultivar lineage is not specified in the text (it is recorded only as 『夏ミカン』 (in translation: "natsumikan")).
For that natsumikan bitterness, debittering treatment using naringinase has advanced as far as factory field trials, and an effective range has been calculated as an addition amount per volume of juice.
Citrus natsudaidai — perceived bitterness is suppressed by: Treatment by adding naringinase (an enzyme that breaks down naringin). A formula was developed to predict the activity needed from pH, temperature, and glucose concentration, and an addition of 50–100 ten-thousand units of enzyme (roughly half a million to a million units) per 1 ton of juice was found to be the effective amount at which the measured and calculated values agreed.supported
Scope: Enzyme treatment experiments and taste tests (including factory field trials) using natsumikan first-press and second-press juice
Note: This is a source that reached sensory evaluation, not just chemical analysis, by directly measuring the time to bitterness disappearance through taste tests (sensory). It also reports that if the reaction time is too long, discoloration, aroma deterioration, and pulp separation occur.
Experimentally formulated the effects of pH, temperature, and glucose concentration, and created a formula to predict the time needed for the juice's bitterness to disappear. Bitterness disappearance was also confirmed by taste tests in factory field trials. An addition of 50–100 ten-thousand units of enzyme (roughly half a million to a million units) per 1 ton of juice was most effective.
Limitations: It is stated that if the reaction time is too long, discoloration, aroma deterioration, and pulp separation occur, leaving a difficult-to-reconcile relationship between enzyme amount and reaction time.
Show dataingredient-natsumikan —has_perceived_bitterness_suppressed_by→ "Treatment by adding naringinase (an enzyme that breaks down naringin). A formula was developed to predict the activity needed from pH, temperature, and glucose concentration, and an addition of 50–100 ten-thousand units of enzyme (roughly half a million to a million units) per 1 ton of juice was found to be the effective amount at which the measured and calculated values agreed."
Kawachi bankan was classified, in a comparative survey of 8 citrus cultivars from Kumamoto Prefecture, into the 「苦味系」 (in translation: "bitter group") that contains abundant naringin.
Scope: Kumamoto Prefectural Industrial Technology Center research, 2005-2007 fiscal years, 8-variety comparison
Kawachi bankan was classified, together with kawano natsudaidai, otachibana, and banpeiyu, into the 『苦味系』 (in translation: "bitter group") that contains abundant naringin in the juice and peel.
Limitations: Quantitative values have not been confirmed (presence/absence classification only).
In the same comparative survey, kawachi bankan was also the only one, among the 8 cultivars tested, whose peel was found to contain nomilin. Including grapefruit, the combination of compounds that carries citrus bitterness differs by species.
Scope: Kumamoto Prefectural Industrial Technology Center research, 2005-2007 fiscal years, 8-variety comparison
Note: This statement concerns the peel (Table 2); it does not address the juice. Nomilin is not recorded for kawano natsudaidai, otachibana, or banpeiyu.
Of the 8 cultivars, nomilin was detected only in kawachi bankan.
Limitations: Quantitative values have not been confirmed (presence/absence classification only).
4. Moved by Processing — Marmalade and Debittering
Even under international standards, marmalade is defined on the premise that it handles bitter material. Codex defines citrus marmalade as a product that may use whole fruit, pulp, juice, peel, and the like.
citrus marmalade is defined in the Codex standard as A product that brings citrus fruit to a suitable viscosity and that may use whole fruit, pulp, purée, juice, water extract, peel, and the likecontextual
Scope: Codex CXS 296-2009 (revised 2025)
Note: A product definition under the standard. It does not indicate the presence or desirability of bitterness.
Context: Defines citrus marmalade as a product that brings citrus fruit to a suitable viscosity and that may use peel and the like.
Limitations: It does not specify sensory characteristics, regional recipes, or degree of liking.
Show dataseasoning-citrus-marmalade —is_defined_in_codex_as→ "A product that brings citrus fruit to a suitable viscosity and that may use whole fruit, pulp, purée, juice, water extract, peel, and the like"
That bitterness in the peel can be moved with alkaline treatment. When the peel of a certain citrus was blanched in NaOH solution and then soaked, naringin in the peel decreased greatly.
Bitterness also moves with salt. In jam made with bitter-orange peel, the water-only treatment did not differ from the bitter control, while the group treated with added NaCl had a lower bitterness rank.
There is also a way of moving it by giving up heat itself. Comparing the same formulation of yuzu-peel marmalade made by a high-pressure method and by a heating method, the high-pressure method was rated as less bitter.
Scope: The same formulation compared under high-pressure processing and heat processing by 27 students and staff
Note: The high-pressure method scored lower on bitterness, and higher on liking of the bitterness and other measures. Not reducible to a high-pressure effect alone.
The high-pressure method was less bitter than the heat method, and scored significantly higher on liking of the bitterness, mouthfeel, and overall evaluation.
Limitations: Details such as blinding are limited. Multiple attributes changed at once.
In grapefruit juice, treatment with a biocatalyst of immobilized naringinase reduced naringin and limonin simultaneously, both falling below the detection threshold.
grapefruit juice can be debittered by: Simultaneous adsorption of limonin with immobilized naringinase on butyl- or octyl-glyoxyl agarose (a 54-74% reduction in naringin, a 76-100% reduction in limonin)supported
Scope: Grapefruit juice treated at pH 4.0 and 30°C. The biocatalyst was reused across 5 batches of 24 hours each
Note: Whether the final concentration reached the cited detection threshold was carrier-dependent: limonin fell below threshold on both carriers, but naringin only on the octyl carrier.
Naringinase immobilized on butyl-glyoxyl agarose lowered naringin and limonin by 54% and 100%; on octyl-glyoxyl agarose, by 74% and 76%. Limonin fell below the detection threshold on both carriers, but naringin only on the octyl carrier. The biocatalyst was reused across 5 batches of 24 hours each.
Limitations: No sensory panel; the bitterness reduction is inferred from literature thresholds; the initial juice concentration varied between batches.
Show dataingredient-grapefruit-juice —can_be_debittered_by→ "Simultaneous adsorption of limonin with immobilized naringinase on butyl- or octyl-glyoxyl agarose (a 54-74% reduction in naringin, a 76-100% reduction in limonin)"
Moving away from the language of processing and analysis, and lining up the language of product descriptions instead, a different line appears. Business-use hassaku juice explicitly states 「独特のほろ苦さ」 (in translation: "a distinctive gentle bitterness") as a feature.
Hassaku is presented as: "The juice has a moderate sweetness and acidity, characterized by a distinctive slight bitterness"observation
Scope: Product description on the official page of Ito Farm (Wakayama Prefecture), a business-use seller of hassaku juice
Note: The same page also explains a production method in which only the flesh is gently squeezed with a bowl-shaped device, so that components from the outer and inner peel are not included in the juice. The relationship between this peel-excluding method and the source of the described slight bitterness (whether from the flesh itself or from residual peel) is not made explicit on the page.
Describes hassaku juice as "having a moderate sweetness and acidity, characterized by a distinctive slight bitterness," and also records a production method in which only the flesh is gently squeezed with a bowl-shaped device so that components from the outer and inner peel are not included in the juice.
Limitations: The relationship between the peel-excluding production method and the source of the described slight bitterness (whether from the flesh itself or from residual peel) is not made explicit on the page.
By contrast, the official product descriptions for mass-market Sunkist's main citrus juices do not use bitterness vocabulary at all.
Japanese citrus beverages is documented as having The word "bitterness" does not appear at all in the official product descriptions of Morinaga Milk Industry's Sunkist mainstream 100% citrus juices (100% Orange, 100% Grapefruit Blend, 100% Orange Blend).observation
Scope: Morinaga Milk Industry's official website, product pages (retrieved 2026-09-04)
Verbatim: "An orange juice with a good balance of sourness and sweetness." No bitterness-related word appears.
Limitations: The description of 1 product; cannot be generalized to other orange beverages.
Verbatim: "A 100% juice beverage blending pink grapefruit, apple, white grape, and lemon, offering a taste like freshly squeezed fruit." No bitterness-related word appears.
Limitations: A fruit blend, not single-variety grapefruit juice.
Show dataconcept-japan-citrus-beverage —is_documented_as_having→ "The word "bitterness" does not appear at all in the official product descriptions of Morinaga Milk Industry's Sunkist mainstream 100% citrus juices (100% Orange, 100% Grapefruit Blend, 100% Orange Blend)."
claim: clm-w61b-sunkist-lineup-no-bitter / type: cultural / updated: 2026-09-04
Even among orange-type products, Natchan Orange places bitterness as something to be suppressed, as 「抑えた飲みやすさ」 (in translation: "easy drinkability with it held down").
Japanese citrus beverages is documented as having Suntory Natchan Orange's official product information describes it as achieving "both the fresh juiciness natural to the fruit and easy drinkability with sourness and bitterness held down," positioning bitterness as something to suppress.observation
Scope: Suntory's official product information page (retrieved 2026-09-04)
Verbatim: "Using a blend juice of orange, known for its refreshing aroma, and mandarin orange, a relative of the mikan familiar to Japanese people, it achieves both the fresh juiciness natural to the fruit and easy drinkability with sourness and bitterness held down."
Limitations: The description of 1 425 ml PET-bottle product.
Show dataconcept-japan-citrus-beverage —is_documented_as_having→ "Suntory Natchan Orange's official product information describes it as achieving "both the fresh juiciness natural to the fruit and easy drinkability with sourness and bitterness held down," positioning bitterness as something to suppress."
claim: clm-w61b-nacchan-bitter-suppressed / type: cultural / updated: 2026-09-04
Yet once it comes to kawachi bankan or hassaku, direct-from-producer growers speak of bitterness positively. Itonoen describes kawachi bankan as 「さっぱりとしたほろ苦さ」 (in translation: "a refreshing, gentle bitterness") and hassaku as 「ほろ苦さがクセになる」 (in translation: "a gentle bitterness that becomes addictive").
Japanese citrus beverages is presented as: The official online shop of Itonoen, a grower in Arida, Wakayama Prefecture, describes kawachi bankan as "also called Japanese-style grapefruit, with a refreshing hint of bitterness," and hassaku as "hassaku, whose hint of bitterness is addictive," presenting bitterness as an active selling point for some cultivars.observation
Verbatim: "You can choose your set contents from the standard satsuma, hassaku with a hint of bitterness that's addictive, or unusual, premium citrus such as shiranuhi, setoka, and sanbokan."
Limitations: Not a description on a dedicated hassaku product page, but a general blurb inside a product-selection guide.
Show dataconcept-japan-citrus-beverage —is_presented_as→ "The official online shop of Itonoen, a grower in Arida, Wakayama Prefecture, describes kawachi bankan as "also called Japanese-style grapefruit, with a refreshing hint of bitterness," and hassaku as "hassaku, whose hint of bitterness is addictive," presenting bitterness as an active selling point for some cultivars."
claim: clm-w61b-itonoen-bitter-value / type: cultural / updated: 2026-09-04
Muchachaen's hassaku juice goes further still, stating that 「独特の苦みこそが八朔らしさ」 (in translation: "it is precisely that distinctive bitterness that makes hassaku what it is").
Japanese citrus beverages is presented as: Muchachaen's hassaku juice states 『ほんのりとした苦味が魅力的な八朔』 (in translation: "hassaku, whose charm is its gentle hint of bitterness") and 『独特の苦みこそが八朔らしさ』 (in translation: "it is precisely that distinctive bitterness that makes hassaku what it is"), emphasizing bitterness most explicitly as a value word among the company's citrus juices.observation
Verbatim: 『ほんのりとした苦味が魅力的な八朔。瀬戸内海の因島で発見され、文旦系などの交雑品種と言われています。（中略）独特の苦みこそが八朔らしさ、昔なつかしい風味をお楽しみください。』 (in translation: "Hassaku, whose charm is its gentle hint of bitterness. It was discovered on Innoshima in the Seto Inland Sea, and is described as a hybrid cultivar of the pomelo lineage, among others. [omission] It is precisely that distinctive bitterness that makes hassaku what it is — please enjoy its nostalgic flavor.")
Show dataconcept-japan-citrus-beverage —is_presented_as→ "Muchachaen's hassaku juice states 『ほんのりとした苦味が魅力的な八朔』 (in translation: "hassaku, whose charm is its gentle hint of bitterness") and 『独特の苦みこそが八朔らしさ』 (in translation: "it is precisely that distinctive bitterness that makes hassaku what it is"), emphasizing bitterness most explicitly as a value word among the company's citrus juices."
claim: clm-w61b-muchachaen-hassaku-bitter-strongest / type: cultural / updated: 2026-09-04
Even lining up multiple producers side by side for the single cultivar kawachi bankan, the wording points in the same direction. Muchachaen, Itonoen, and the JA Kumamoto Fruit Growers' Association Federation all described bitterness as a value word.
Japanese citrus beverages is documented as having Regarding juice of the single cultivar kawachi bankan, the direct-from-producer sellers Muchachaen, Itonoen, and the JA Kumamoto Fruit Growers' Association Federation all describe bitterness as a value word — 『心地よい苦み』 (in translation: "a pleasant bitterness"), 『ほろ苦さ』 (in translation: "a gentle bitterness"), 『大人の味』 (in translation: "a grown-up flavor") — and this direction agrees with the wording of Morinaga Milk Industry's mass-market Sunkist 『濃いグレープフルーツ』 (in translation: "Rich Grapefruit"), described as having 『ほのかな苦み』 (in translation: "a faint bitterness"). On the other hand, even on the same mass-market side, Suntory Natchan describes bitterness as something 『抑えた』 (in translation: "held down"), and Kagome's limited-edition 「温州しぼり」 (in translation: "Unshu Shibori") also treats bitterness as something to be excluded, in its explanation of the pressing method.observation
Scope: Official sites of Muchachaen, Itonoen, JA Kumamoto Fruit Growers' Association Federation, Morinaga Milk Industry, Suntory, and Kagome (accessed 2026-09-04; Morinaga Milk Industry only, 2026-09-03)
Note: Verification of the H10 reformulation: what separates bitterness as a value word from bitterness as something to be suppressed is not the dichotomy of 『大量流通／差別化』 (in translation: "mass-market vs. differentiated"), since even within the single cultivar kawachi bankan there are examples on both the mass-market and differentiated sides that treat bitterness as a value word, and conversely the mass-market side itself contains both suppressive and affirmative wording. Rather than defend the hypothesis, only the observation itself is recorded.
Verbatim: 『その後感じる心地よい苦みがさっぱりとした余韻を残し』 (in translation: "the pleasant bitterness felt afterward leaves a refreshing finish") (differentiated; kawachi bankan).
Verbatim: 『和製グレープフルーツとも呼ばれる、さっぱりとしたほろ苦さ。』 (in translation: "a refreshing, gently bitter flavor, also called a Japanese-made grapefruit.") (differentiated; kawachi bankan).
Verbatim: 『程よい酸味のきいたほろ苦さは、まさに"大人の味"。』 (in translation: "a gentle bitterness with just the right tartness — truly a 'grown-up flavor.'") (differentiated; kawachi bankan).
Contradicts: Verbatim: 『酸味や苦味を抑えた飲みやすさを両立しました。』 (in translation: "we achieved easy drinkability by holding down sourness and bitterness.") (mass-market; orange type; suppressive wording).
Limitations: The cultivar is an orange type, different from grapefruit or kawachi bankan.
Contradicts: Verbatim: 『果皮の苦味や渋みが混ざらないように...インライン搾汁』 (in translation: "...in-line pressing, so that the bitterness and astringency of the peel do not mix in...") (a limited-edition product from a mass-market maker; suppressive wording).
Limitations: The cultivar is unshu mikan, different from grapefruit or kawachi bankan.
Show dataconcept-japan-citrus-beverage —is_documented_as_having→ "Regarding juice of the single cultivar kawachi bankan, the direct-from-producer sellers Muchachaen, Itonoen, and the JA Kumamoto Fruit Growers' Association Federation all describe bitterness as a value word — 『心地よい苦み』 (in translation: "a pleasant bitterness"), 『ほろ苦さ』 (in translation: "a gentle bitterness"), 『大人の味』 (in translation: "a grown-up flavor") — and this direction agrees with the wording of Morinaga Milk Industry's mass-market Sunkist 『濃いグレープフルーツ』 (in translation: "Rich Grapefruit"), described as having 『ほのかな苦み』 (in translation: "a faint bitterness"). On the other hand, even on the same mass-market side, Suntory Natchan describes bitterness as something 『抑えた』 (in translation: "held down"), and Kagome's limited-edition 「温州しぼり」 (in translation: "Unshu Shibori") also treats bitterness as something to be excluded, in its explanation of the pressing method."
claim: clm-w61b-kawachibankan-cross-value-h10 / type: cultural / updated: 2026-09-04
That said, the language around kawachi bankan does not all agree. Ainan Town's introduction of the region's produce describes the taste as like 「苦味のない和製グレープフルーツ」 (in translation: "a Japanese-made grapefruit with no bitterness"), while analysis by the Kumamoto Prefectural Industrial Technology Center detected naringin, limonin, and nomilin, and classified it among the cultivars tested as part of the bitter group. Chemical analysis and the producing region's narrative can point in opposite directions about the same fruit.
Kawachi bankan is presented as: It is described, in the producing region's introductory text, as tasting like 『ほどよい酸味の効いた、苦味のない和製グレープフルーツ』 (in translation: "a Japanese-made grapefruit with just the right tartness and no bitterness").contextual
Scope: Product introduction text in official materials from Ainan Town, Ehime Prefecture
After stating 『文旦から派生したといわれており糖度は約11％』 (in translation: "it is described as derived from pomelo, with a sugar content of about 11%"), it introduces the fruit as 『ほどよい酸味の効いた「苦味のない和製グレープフルーツ」のような味わい』 (in translation: "a taste with just the right tartness, like a 'Japanese-made grapefruit with no bitterness'").
Limitations: It is not specified whether this description is based on sensory evaluation or is simply a promotional impression.
Contradicts: For the same kawachi bankan, analysis by the Kumamoto Prefectural Industrial Technology Center detected naringin, limonin, and nomilin, positioning it, among the 8 cultivars tested, as belonging to the 『苦味系』 (in translation: "bitter group") and as the only one that also carries nomilin.
Limitations: This is a discrepancy between the chemical analysis and the producing region's promotional narrative; this survey cannot judge which description better reflects the actual sensory reality.
Show dataingredient-kawachi-bankan —is_presented_as→ "It is described, in the producing region's introductory text, as tasting like 『ほどよい酸味の効いた、苦味のない和製グレープフルーツ』 (in translation: "a Japanese-made grapefruit with just the right tartness and no bitterness")."
claim: clm-w40b-kawachibankan-presented-bitterless / type: cultural / updated: 2026-09-02
What is not known
Whether acid or an enzyme drives limonin's delayed bitterness has not been settled between separate sources — kinetics from seedless juice on one side and purification of a seed-derived enzyme on the other.
The receptor-level basis of naringin also remains unresolved. Naringin does not activate TAS2R19 in vitro, and the receptor suggested by grapefruit-liking genetics was verified with quinine, not with naringin.
Some processed citrus products, such as chinotto marmalade, appear in the encyclopedia's item list but could not be traced to direct measurement or observation. Rather than putting the name forward on its own, this is left blank until the next source is found.
Frequently asked
Why does squeezed orange juice grow more bitter over time?
It has been confirmed that this happens because the non-bitter precursor (LARL) present in freshly squeezed orange juice converts into bitter limonin while in storage. Direct measurements using navel orange juice show that limonin content rises during storage after pressing and eventually reaches a constant value. Whether the mechanism driving this change is a chemical reaction promoted by acid or one driven by a seed-derived enzyme is explained differently between studies, and has not yet been confirmed. It has also been reported that heat processing and acidification only speed up this conversion; they do not change the upper limit itself of how bitter it eventually becomes.
Is naringin alone responsible for grapefruit's bitterness?
Naringin and limonin have long been treated as the two major compounds involved in grapefruit's bitterness, but which bitter receptor actually senses naringin has not been settled. Naringin does not activate the receptor TAS2R19 in vitro, and TAS2R31, the receptor suggested by grapefruit-liking genetics, has only been confirmed with quinine, not verified with naringin itself. Taste-reconstitution experiments using orange juice likewise found limonin, limonoid glycosides, polymethoxyflavones, acids, and sugars intertwined in complexity, with no single indicator found that explains bitterness. In other words, that naringin itself is bitter has been confirmed, but that alone does not explain the bitterness of citrus as a whole.