Dill

What is Dill?

Dill (Anethum graveolens) seed oil was in William Woodward's original gripe water, and the UK product licensed today carries 2.3 mg of it per 5 mL spoonful. That figure comes from the Summary of Product Characteristics for Woodward's Gripe Water (Alcohol Free and Sugar Free), marketing authorization PL 00063/0698, held by Reckitt Benckiser Healthcare (UK) and last revised 14 October 2025. Each 5 mL also holds 52.5 mg of sodium hydrogen carbonate, which is to say baking soda, and the authorized indication is "symptomatic relief of distress associated with wind in infants up to one year old."

The plant itself is an annual of the carrot and parsley family, Apiaceae, usually 40 to 60 cm tall, with threadlike feathery leaves and small yellow flowers in umbels 2 to 9 cm across. Two different things are harvested from it. The leaf, sold fresh or dried as dill weed, is a culinary herb. The dried fruit, sold as dill seed, is the spice and the source of the pharmacopeial drug. Their volatile oils barely overlap. Charles, Simon and Widrlechner characterized germplasm from the US National Plant Germplasm System in the Journal of Essential Oil Research in 1995 and found alpha-phellandrene at 51.1 to 64.7 percent of herb oil with no carvone detected, while seed oil was 68 to 83 percent carvone plus dihydrocarvone, 14.18 to 21.43 percent limonene, and no dill ether at all.

Common Names

• Dill, dill weed (the leaf), dill seed (the fruit)

• Anethi fructus, the pharmacopeial Latin for the dried fruit

• Peucedanum graveolens, the older botanical name used by the British Pharmaceutical Codex of 1911

• Indian dill, sowa, Anethum sowa: the South Asian type, often treated as a separate species and chemically distinct

• Shevid, the Persian name; nearly every human dill trial cited in this issue was run in Iran

Primary Active Compounds

• (+)-Carvone, the dominant ketone of the seed oil: 40 to 60 percent under the 1911 British Pharmaceutical Codex monograph, 50 to 60 percent in US-grown and 35 to 60 percent in European oils per the PROSEA plant resources handbook. Its pharmacology is covered in this archive's caraway issue.

• d-Limonene, the main hydrocarbon of seed oil, 14 to 21 percent in the Charles survey

• Alpha-phellandrene, the dominant constituent of herb oil

• Dill ether (3,9-epoxy-p-menth-1-ene), about 25 percent of herb oil per PROSEA and one of the main odorants of the fresh leaf

• Dillapiole, a phenylpropanoid that reached 20.7 percent of Indian dill seed oil in Lawrence's 1980 analysis and is largely absent from European dill

• Furanocoumarins: oxypeucedanin, oxypeucedanin hydrate and a previously undescribed linear furanocoumarin isolated from the whole herb by Stavri and Gibbons in 2005, plus bergapten reported in the fruit

Key Note

Germany's Commission E approved dill fruit in 1987 for dyspeptic complaints, an indication that says nothing about infants, and the European Medicines Agency's herbal committee has no monograph for dill fruit. The infant use that made dill famous rests on a Victorian proprietary formula, not on any regulatory review of the herb.

Dill is a real spice with a modest adult evidence file and a famous infant use that the evidence never reached.

Primary Functions & Benefits

Dill's defensible adult categories are digestive (the Commission E indication), menstrual pain (small trials, the best reported of them against an under-dosed comparator), and lipids (contradictory). Its infant category is historical.

What regulators have said

Commission E's 1987 monograph approved dill fruit for dyspeptic complaints and listed no known contraindications. The UK's licensed gripe water rests on a traditional formula rather than a trial: its SmPC describes dill seed oil as an aromatic carminative used for flatulence in children and credits sodium hydrogen carbonate with "a well-established antacid action."

Gut symptoms, measured in people

Only one adult trial has scored digestive symptoms. Haidari and colleagues randomized 48 adults with type 2 diabetes and gastrointestinal complaints to 1 g capsules of dried dill leaf powder three times daily or starch for 8 weeks; 42 finished. Of five symptom categories, only colonic motility disorders fell significantly (p = 0.01). Dyspepsia, reflux, esophageal motility and gastric motility did not change.

Topical and local antimicrobial use

Saghafi and colleagues randomized 60 women with culture-confirmed vulvovaginal candidiasis to 2 percent dill vaginal suppositories or 100 mg clotrimazole vaginal tablets in 2018. Positive cultures at the end were 13 percent with dill and 10 percent with clotrimazole (p = 0.68). A single-blind, single-center equivalence claim from 60 women is a pilot.

Forms & Standardization

Dill is sold as seven materials, and the infant products add an eighth that may contain none of the plant.

Whole dried fruit (dill seed)

Volatile oil content is 2 to 6 percent on a dry basis per PROSEA, and 2.5 to 4.0 percent in the German Medpharm monograph. Harvest stage moves it: PROSEA reports 2.6 to 3.7 percent in green fruit, falling to about 1.7 percent once the fruit changes color. The Commission E dose of 3 g of fruit therefore carries roughly 75 to 120 mg of oil at 2.5 to 4 percent, which lands at or just below the bottom of the monograph's 0.1 to 0.3 g oil range.

Dried dill weed

The leaf, specified for trade by ISO 21803:2019, "Dried dill: Specification." Herb oil content is 0.1 to 1.5 percent per PROSEA, an order of magnitude below the fruit.

Dill seed oil (European type)

Steam distilled from the fruit. The 1911 British Pharmaceutical Codex set specific gravity at 0.900 to 0.915, optical rotation at +70 to +80 degrees and carvone at 40 to 60 percent, and warned that a specific gravity under 0.900 may mean carvone had been extracted. Modern carvone figures run 50 to 60 percent in US oils and 35 to 60 percent in European ones (PROSEA).

Terpeneless dill seed oil

The grade in the UK gripe water. Removing the terpene hydrocarbons, chiefly limonene, leaves an oil more concentrated in carvone. The SmPC states the oil grade and the 2.3 mg per 5 mL amount; it does not publish a carvone percentage.

Dill weed oil

Distilled from the flowering herb. Alpha-phellandrene, beta-phellandrene and dill ether together made up 90 to 97 percent of herb oil in the Charles survey. A supplement or aromatherapy product that says only "dill oil" has not told you which of two chemically opposite oils it contains.

Indian dill seed oil (Anethum sowa)

Lawrence's 1980 analysis, as tabulated by PROSEA, found limonene 45.0 percent, carvone 23.1 percent, dillapiole 20.7 percent, and cis- and trans-dihydrocarvone 5.2 and 4.2 percent. The 1911 Codex already separated it from European oil by its lower rotation (+41 to +47 degrees) and higher specific gravity (0.948 to 0.975), noting that its dillapiole sinks in water. Dillapiole is an isomer of the apiole covered in this archive's parsley issue. Human toxicology data on dillapiole itself are thin, so the useful buyer instruction is to know which dill oil is in the bottle.

Dill water

The historic pharmacy vehicle: the 1911 Codex's concentrated dill water was diluted one part to 40. The archived earlier revision of FDA's import alert names dill water alongside dill oil.

Gripe water, by jurisdiction

• UK, Woodward's (Alcohol Free and Sugar Free), first authorized 30 November 1999: beyond the dill oil and bicarbonate, each 5 mL carries 950 mg maltitol, 200 mg propylene glycol and 15.73 mg sodium.

• India, Woodward's Gripe Water (TTK Healthcare, sold as an Ayurvedic formulation): terpeneless dill seed oil, sarjikakshara (sodium bicarbonate) and sugar. Its directions start at 0 to 1 month, an age the UK authorization excludes outright.

• United States: the three products in the previous section, none listing dill. Beyond their herbs, Mommy's Bliss adds glycerin, citrus bioflavonoid extract and potassium sorbate, Little Remedies adds agave syrup and glycerin, and Colic Calm's homeopathic list includes vegetable charcoal.

Mommy's Bliss label directions allow up to 15 servings a day for babies 2 to 6 weeks old. No trial of that product, at that frequency, in that age band, appears in the literature reviewed for this issue.

Capsules and powders used in adult trials

Seed powder, dried leaf powder, a hydroalcoholic leaf-and-stem extract, and two tablets of undisclosed composition, each listed with its trial under Recommended Dosages. Mansouri's group noted that the earliest human lipid trials had used a patented multi-herb product that was 68 percent dill. None of the trial preparations reported a carvone or other marker content, and no dill supplement sold in the US is standardized in a way that matches any trial.

Food Sources

Fresh dill weed is a food with real micronutrients at 100 g and trivial ones at the amount anyone eats.

Fresh dill weed, per 100 g (USDA FoodData Central, SR Legacy 172233)

43 kcal, 86 g water, 3.46 g protein, 2.1 g fiber, 208 mg calcium, 6.59 mg iron, 55 mg magnesium, 738 mg potassium, 1.26 mg manganese, 85 mg vitamin C, 386 micrograms vitamin A as retinol activity equivalents, and 150 micrograms folate as dietary folate equivalents.

Fresh dill weed, per realistic serving

USDA's one-cup-of-sprigs serving weighs about 9 g and delivers 18.5 mg calcium, 0.59 mg iron, 7.6 mg vitamin C and 13.4 micrograms folate. That cup is a large garnish. Its vitamin C is about 8 percent of the 90 mg adult male requirement, and its iron is under 4 percent of an adult woman's 18 mg.

Dill seed, per 100 g (USDA FoodData Central, SR Legacy 170925)

305 kcal, 16 g protein, 14.5 g fat, 21.1 g fiber, 1,516 mg calcium, 16.3 mg iron, 256 mg magnesium, 1,186 mg potassium, 1.83 mg manganese, 21 mg vitamin C. The calcium figure is striking and almost irrelevant: at the Commission E dose of 3 g, it is about 45 mg.

Seed versus gripe water, by oil content

The 2.3 mg of dill seed oil in one 5 mL spoonful of the UK gripe water equals the oil in about 60 to 90 mg of whole seed at 2.5 to 4 percent oil. The infant product's dill is a flavoring quantity, and the comparison is not a reason to give seed or oil to a baby.

Is dietary intake enough?

There is no dill requirement to meet. The Commission E amount, 3 g of crushed seed, is a culinary quantity, so food is the form that matches the monograph.

Who Should Take Dill

No infant should be given dill for colic on the strength of the evidence, and most adults have no reason to take it beyond the spice rack.

Adults for whom a trial of dill is reasonable

• Adults with uncomplicated, already-evaluated dyspepsia or flatulence who want the Commission E use. This is a traditional-use approval, and the one adult trial that scored dyspepsia found no change.

• Women with primary dysmenorrhea who want a non-drug option to trial for two cycles, understanding that the comparison drug in the best-reported trial was given below its labeled dose.

• Cooks. Dill is safe and pleasant as food for nearly everyone without an Apiaceae allergy.

Who should be evaluated instead of dosed

Parents of a crying infant. Infantile colic is self-limiting, and the UK's National Institute for Health and Care Excellence, in its Clinical Knowledge Summary as quoted by the NHS-funded prescribing support organization PrescQIPP in Bulletin 303 (July 2022), holds that there is insufficient good-quality evidence to support prescribing or recommending any medication for it. PrescQIPP advises reviewing and stopping prescriptions of gripe water, simeticone and lactase drops for colic and not recommending herbal supplements. The useful act is a clinical check that the crying is colic, followed by time.

Who Should AVOID or Use Caution

Contraindications

• Known allergy to dill or cross-reacting Apiaceae spices. Chiu and Zacharisen at the Medical College of Wisconsin reported in 2000, under the title "Anaphylaxis to dill," what they described as the first confirmed dill allergy: a woman with allergic rhinitis developed oral itching, tongue and throat swelling, urticaria, vomiting and diarrhea after eating food cooked with dill and later on inhaling it during cooking, with a positive skin test to fresh dill. Freeman described allergy to fresh dill in Allergy in 1999. In a Málaga aniseed allergy case, García-González and colleagues showed IgE cross-reactivity among aniseed, caraway, coriander, fennel and dill extracts by enzyme immunoassay inhibition.

• Any gripe water containing sodium bicarbonate in infants with impaired kidney function, metabolic or respiratory alkalosis, low calcium or low chloride. Those are the UK SmPC's own contraindications for the licensed dill product, which also excludes infants hypersensitive to its hydroxybenzoate (paraben) preservatives and warns they may cause delayed allergic reactions.

• Infants under one month, for the UK dill product. Its posology reads "Not to be used" for that age.

• Hereditary fructose intolerance, for gripe waters sweetened with maltitol, per the same SmPC.

Use Caution

• Pregnancy. Dill seed decoction is studied in Iran specifically as a way to shorten labor, which is a reason not to take medicinal amounts before term. Hekmatzadeh and colleagues gave a single dill seed decoction at the start of active labor, in hospital, in a 153-participant trial against routine care and reported shorter labor stages in the dill group. Monsefi and colleagues fed female Wistar rats high-dose dill extracts in 2006 and reported a longer estrous cycle and higher progesterone.

• Heavy menstrual bleeding. In Heidarifar's dysmenorrhea trial, 2 of 23 women in the dill arm reported heavier or longer bleeding, against 1 in each comparison arm. That difference was not significant and the numbers are tiny, which is why this sits under caution rather than contraindication.

• Photosensitizing medication or PUVA therapy with methoxsalen. Dill contains linear furanocoumarins, and Egan and Sterling listed dill among common photosensitizing vegetables in a 1993 Cutis report on phytophotodermatitis. The risk from eating the herb is low; the risk from handling large amounts of the fresh plant in sun, or from concentrated topical preparations, is the realistic one.

• Glucose-lowering medication. Dill trials in type 2 diabetes measured insulin and insulin resistance; that territory belongs to another issue in this batch and is flagged here only as a reason to monitor.

Critical Safety Point

The highest-stakes use of dill is the one with the least evidence: a bottle given to a newborn. Every infant product number in this issue is a manufacturer's direction or a regulator's limit on one formula, not evidence that dill helps a baby. A US gripe water has no FDA review behind its directions, and three of the infant gripe water incidents documented in the quality section (a Cryptosporidium contamination, a septic shock case and an undissolved-ingredient recall reported in a one-week-old) happened to products sold for exactly this purpose. Infant feeding guidance from the WHO and UNICEF Baby Friendly Hospital Initiative, as Adhisivam summarized in 2012, is that newborns be given no food or drink other than breast milk unless medically indicated. A baby who is feeding poorly, vomiting, febrile, or crying in a way that is new or different needs to be seen, not soothed.

Recommended Dosages

No dose of dill for an infant is recommended anywhere in this issue, because none is supported. The numbers below are adult figures: a German regulatory monograph and the preparations used in specific adult trials. None of them is evidence for use in babies, and none describes a dill oil dose anyone should measure out at home.

Culinary dill (adults)

No limit of concern for an adult without Apiaceae allergy. Dill weed and dill seed are foods.

Dill fruit, Commission E (adults, dyspeptic complaints)

3 g of the dried fruit per day on average, from the 1987 monograph. That is an ordinary kitchen quantity of crushed seed, and the monograph's figure is a traditional-use posology rather than a dose established in a dose-finding trial.

Dill essential oil

No amount is given here for anyone. The monograph's essential oil figure appears under Forms & Standardization only as a measure of how much oil 3 g of fruit carries; no human oral tolerance study of isolated dill oil turned up in the research for this issue.

Preparations used in adult trials

• Dysmenorrhea: dill seed powder, two 500 mg capsules every 12 hours (2 g a day) for 5 days starting 2 days before menstruation, over two cycles. Heidarifar 2014, 75 randomized, 70 analyzed, 23 in the dill arm.

• Type 2 diabetes with digestive symptoms: dried dill leaf powder, 1 g three times daily after meals (3 g a day) for 8 weeks. Haidari 2020, 48 randomized, 42 analyzed.

• Hyperlipidemia: "anethum tablet" 650 mg twice daily for 6 weeks. Kojuri 2007, 150 randomized, 50 per arm. Null result.

• Metabolic syndrome: 600 mg hydroalcoholic extract of dried leaves and stems once daily for 12 weeks. Mansouri 2012, 24 randomized, 20 analyzed.

• Hypercholesterolemia: six Iran Daruk dill tablets daily, content unstated, for 2 months. Mirhosseini 2014, 91 allocated, 35 in the dill arm completed.

• Vaginal candidiasis: 2 percent dill vaginal suppositories. Saghafi 2018, 60 women.

The labor trials used a seed decoction in women already in active labor under obstetric care. Those preparations appear under Special Considerations, not here, because they are not doses for anyone to take at home.

Duration

Adult trial exposures ran from 5 days per cycle to 12 weeks. Nothing longer than 12 weeks of oral dill has been tested in a controlled human study cited here.

Timing & Administration

With or after meals for digestive use. The Commission E indication is postprandial discomfort, and the Haidari capsules were taken after breakfast, lunch and dinner.

Before the period, not during it, for cramps. Heidarifar started dill 2 days before expected menstruation and continued for 5 days. Starting once pain has peaked was not tested.

Crush seed just before use. The oil percentages in the forms section were measured on intact fruit, and the oil is volatile once the fruit is broken, so ground dill seed bought months earlier is a weaker version of the Commission E material. No pharmacopeia cited here sets a dill seed infusion strength.

Fresh herb near the end. The odorants that define dill weed (alpha-phellandrene, limonene, dill ether, myristicin, per Jana and Shekhawat's 2010 review) are small volatile molecules, which is the practical case for adding the leaf late or raw.

Timeline of Effects

In infants, there is no dill timeline. No controlled study found in this research timed a dill product's effect on infant crying. The one-week result in the herbal colic literature comes from a fennel, chamomile and lemon balm blend, described under Benefits, and it does not transfer to dill or to any US gripe water.

One to two cycles, for menstrual pain. In Heidarifar's trial, pain scores fell significantly from baseline by the first cycle in both the dill and mefenamic acid arms, while placebo reached significance only by the second cycle. The between-group comparison of dill against mefenamic acid showed no difference at either month. The comparator was 250 mg of mefenamic acid every 12 hours, while the US Ponstel label for primary dysmenorrhea specifies 500 mg initially then 250 mg every 6 hours. Dill matched a drug given at 500 mg a day, half the labeled maintenance amount of 1,000 mg a day.

Six to twelve weeks, for lipids, with the effect disappearing under control. Kojuri's 6-week trial found nothing: triglycerides rose 14.74 mg/dL in the dill arm, and total and LDL cholesterol changes were not significant. Mansouri found no change at 8 weeks; at 12 weeks triglycerides fell 55.44 mg/dL from baseline on dill (p less than 0.01) and 55.7 mg/dL on placebo (not significant within that arm), with no difference between arms. The within-group p value is the one the abstract leads with.

When to conclude it is not working. For cramps, two cycles is the trial length; no change by then is an answer. For lipids, a single panel at 8 to 12 weeks would show the effect if it existed, but dill is not a substitute for a statin or fibrate in anyone whose clinician has prescribed one. For infant crying, the timeline is the condition's own: colic is self-limiting, which is precisely why uncontrolled before-and-after impressions of gripe water are uninterpretable.

Benefits of Taking Dill

Primary dysmenorrhea, modest and low quality. Heidarifar's Qom trial enrolled single female students aged 18 to 28 with moderate or severe primary dysmenorrhea in 2011. The 2016 Cochrane review of dietary supplements for dysmenorrhea (Pattanittum and colleagues, 27 trials, 3,101 women) pooled this as one trial and found "no consistent evidence of effectiveness" for dill against placebo (mean difference 1.15 points lower on a 10-point scale, 95 percent CI 2.22 to 0.08 lower, 46 women) and no difference against an NSAID (0.13 points, 95 percent CI 1.01 lower to 1.27 higher, 47 women). It rated all the evidence low or very low quality. A 2026 meta-analysis of Apiaceae herbs for dysmenorrhea by Thaliyil Poduval and colleagues found a significant pooled effect for dill and fennel seed against mefenamic acid or ibuprofen, while flagging concerns about deviations from intended interventions and selective reporting in exactly those dill and fennel versus standard-treatment trials.

Blood lipids, contradictory. Two meta-analyses published in 2021 reached different answers from overlapping trials. Mousavi and colleagues pooled 7 trials with 330 participants and found LDL cholesterol 15.64 mg/dL lower (95 percent CI 24.55 to 6.73 lower) with no significant effect on total cholesterol, triglycerides or HDL. Sadeghi and colleagues pooled 6 trials with 171 intervention participants and reported significant reductions in total cholesterol, LDL and triglycerides. Underneath them, beside the null Kojuri and Mansouri trials: Haidari's 42 completers showed LDL falling 9.76 mg/dL on dill against a 3.09 mg/dL rise on placebo, and Mirhosseini's positive 18 percent total cholesterol drop came from a trial that assigned patients alternately by order of arrival rather than by a random sequence, measured no LDL, and compared dill against gemfibrozil with no placebo arm.

Digestive comfort, by tradition plus one partial signal. The Commission E approval stands; the single controlled human signal is Haidari's colonic motility finding, from a trial whose primary outcome was LDL.

Vaginal candidiasis, one pilot. Equivalence with clotrimazole in 60 women (Saghafi 2018), single-blind.

Infant colic: none for dill. The placebo-controlled infant colic trials of Apiaceae herbs are fennel trials. Alexandrovich and colleagues in St Petersburg randomized 125 infants aged 2 to 12 weeks to a fennel seed oil emulsion or placebo; colic resolved by Wessel criteria in 40 of 62 analyzed on fennel (65 percent) against 14 of 59 on placebo (23.7 percent), with 121 of the 125 contributing to that result. Savino and colleagues in Turin randomized 93 breastfed colicky infants to ColiMil, a standardized extract of Matricaria recutita, Foeniculum vulgare and Melissa officinalis, or placebo for 1 week; 88 completed (41 and 47). Daily crying fell from 201.2 to 76.9 minutes on the extract and from 198.7 to 169.9 minutes on placebo (p less than 0.005). Anheyer and colleagues' 2017 Pediatrics systematic review of herbal medicines for pediatric gastrointestinal disorders (14 trials, 1,927 participants) found evidence for "different fennel preparations" in colic. None of those products is a dill product, and none is a US gripe water.

Breast milk supply: no valid trial. The NIH LactMed record for dill (revised 15 May 2025) calls it "a purported galactogogue" and states that "no scientifically valid clinical trials support this use." The evidence it lists is a nonrandomized, unblinded study of dill seed tea in 60 mothers of neonatal intensive care infants, and a randomized, double-blind trial of Ayush-SS, a multi-herb product containing 1.5 g of dill per dose, after which 70 percent of the treated group and 45 percent of the placebo group were exclusively breastfeeding at 6 months.

Potential Negatives & Side Effects

Allergy, including by inhalation. Celery-mugwort-spice syndrome, the label Wüthrich and colleagues used in Swiss and German reports in 1984 and 1985, links celery allergy and mugwort pollen sensitization with reactions to Apiaceae spices; in Wüthrich and Hofer's 31 celery-allergic patients, clinically relevant reactions to caraway, parsley, fennel and aniseed ran 26, 16, 13 and 3 percent. Anyone with that profile should treat dill as a likely cross-reactor until tested.

Phototoxicity. Ojala and colleagues at the University of Helsinki included dill leaf among six furanocoumarin-bearing Apiaceae (with parsley, angelica and lovage) in a 1999 brine shrimp phototoxicity screen and reported results in line with each plant's furanocoumarin content.

Gastrointestinal and minor complaints. In Heidarifar's trial, one student on dill reported gastrointestinal discomfort. A separate Iranian dysmenorrhea study cited by Heidarifar (Mohammadinia) recorded nausea and dizziness in 6 students taking dill. In Mansouri's 12-week extract trial, a few participants complained about the capsules' smell; laboratory safety panels did not change.

Gripe water, where the excipients carry most of the risk. Sodium bicarbonate given continuously and in large amounts can cause metabolic alkalosis and milk-alkali syndrome, as Adhisivam warned in 2012, adding that hyperacidity is not a cause of colic and the antacid therefore has no role. Sugar-sweetened versions expose erupting teeth. Maltitol, the UK sugar substitute, has a laxative effect noted on the SmPC. And the original product contained alcohol, per Adhisivam: secondary accounts of Blumenthal's 2000 history in the Journal of the Royal Society of Medicine put Woodward's formula at about 3.6 percent, and Adhisivam reported other commercial gripe waters reaching 9 percent. The UK licensed product is now alcohol free; the same accounts date Britain's requirement to remove the alcohol to 1992.

Association with infant symptoms. Jain and colleagues interviewed 335 mothers of infants aged 1 to 6 months at an Indian well-baby clinic in 2015; 64.18 percent were giving gripe water. Colic, vomiting and constipation were all more common in the gripe water group (p = 0.0001, 0.0373 and 0.0007). This was cross-sectional: parents give gripe water to babies who already cry and spit up, so the design cannot separate cause from reason for use. What it does not offer is any support for the idea that routine gripe water prevents colic.

Displacement of breastfeeding. Adhisivam's central objection, framed against step six of the WHO and UNICEF Baby Friendly Hospital Initiative, is that anything other than breast milk given in the first six months can introduce bacteria, delay the establishment of breastfeeding and reduce milk supply.

Deficiency Symptoms

What dill deficiency addresses: no such state exists. No human requirement has been set for carvone, phellandrene, dill ether or any other compound particular to dill, and going without dill produces no syndrome.

The nutrient that dill does contain in quantity, vitamin C at 85 mg per 100 g of fresh leaf, has a real deficiency disease, scurvy, and a defined adult requirement. But that requirement is met by citrus, peppers, potatoes or almost any vegetable, and the 9 g cup of dill sprigs supplies 7.6 mg. Nobody's vitamin C status depends on dill.

Bottom line: dill is optional. A label that implies a body needs it is selling flavor as physiology.

Toxicity Symptoms

At high intake: no toxic syndrome from culinary dill has been characterized in humans, and the largest repeated oral amounts in controlled adult trials were 3 g a day of dried leaf for 8 weeks and 2 g a day of seed powder for 5 days per cycle, and the labor trials gave a single decoction of 10 g of seed in hospital, none with serious adverse events reported. Animal figures exist but describe injected or weight-scaled doses: Hosseinzadeh's group measured intraperitoneal LD50s in mice of 3.04 g/kg for an aqueous seed extract and 6.98 g/kg for an ethanolic one, and Mansouri's group cited animal work by Arbabi and colleagues treating oral hydroalcoholic extract doses below 50 mg/kg as a no-observed-adverse-effect level. Concentrated dill oil is the form with no human oral safety data in the record reviewed here, and Indian dill oil adds dillapiole at about a fifth of its composition.

For infant products, the burden worth counting is the excipients rather than the dill. The UK SmPC calculates that at its own upper limit of directions, a 6 to 12 month old would receive 188.76 mg of sodium and 2,400 mg of propylene glycol in a day, and warns about the sodium for babies on a low-salt diet and about propylene glycol combined with other medicines containing propylene glycol or alcohol.

Signs to reduce or stop: in an adult, hives, lip or throat swelling, wheeze, or vomiting soon after eating or smelling dill (allergy, which can escalate); a blistering or streaky rash in sun-exposed skin that touched the plant (phototoxicity); unusually heavy periods while taking dill capsules. In an infant given any gripe water, stop and get the baby assessed for vomiting, worsening constipation, poor feeding, unusual sleepiness or fast breathing. The last three are not dill side effects; they are reasons a crying baby needs a clinician rather than a bottle.

General note: at food amounts, dill's hazards are allergy and sun-related skin reactions. Most of the documented harm attached to its name came from infant products, several of which contained no dill.

How Dill Works

Spasmolytic action in gut tissue. Gharib Naseri and Heidari added a hydroalcoholic dill fruit extract to Wistar rat ileum in 2007 at 0.5 to 4 mg/mL. It damped contractions driven by potassium chloride, acetylcholine and barium chloride, and it suppressed calcium-induced contractions, a profile the authors read as voltage-dependent calcium channel blockade. That is a tissue-bath mechanism, and it has not been measured in human gut by manometry, breath testing or transit study.

Gastric acid and mucosal protection, in mice. Hosseinzadeh, Karimi and Ameri at Mashhad gave mice aqueous and ethanolic seed extracts in 2002. High oral doses protected the gastric lining against hydrochloric acid and ethanol about as well as sucralfate, and the extracts lowered acidity and total acid output in pylorus-ligated animals. No human gastric secretion study of dill was found.

Why gripe water may seem to work: the sweet taste. Adhisivam suggested in 2012 that any soothing effect of gripe water could come from its sweet taste. Two small controlled studies support that. Markestad in Bergen gave 19 colicky infants a 12 percent sucrose solution or water during crying in a double-blind, double crossover study in 1997; 12 improved specifically on sucrose and 1 on water (p less than 0.01). Barr and colleagues in Montreal compared 19 infants with colic and 19 without in 1999 and found sucrose taste calmed crying in both groups but less effectively in the colicky infants, and that sucrose calming weakens in normal infants by 4 to 6 weeks of age. A sweet liquid will quiet some babies briefly whether or not it contains dill. That is an explanation for parental experience, not a recommendation to give sugar.

Antimicrobial action. Dill seed oil and extracts inhibit bacteria and fungi in vitro, which fits the candidiasis pilot. Stavri and Gibbons found oxypeucedanin, oxypeucedanin hydrate and falcarindiol from the whole herb active against rapidly growing mycobacteria at minimum inhibitory concentrations of 2 to 128 micrograms per mL.

Synergistic Supplements

Dill's best-documented combinations are ones in which dill is either absent from the effective product or one ingredient among several.

Combinations with human data

• Chamomile, fennel and lemon balm (ColiMil): the strongest infant colic result in this issue, with no dill in it.

• Ayush-SS, a multi-herb galactagogue containing 1.5 g dill per dose: the LactMed-cited lactation trial, a mixture result.

• The patented multi-herb lipid product that was 68 percent dill, used in early Iranian lipid trials and described by Mansouri's group as showing no significant lipid effect in humans.

Gripe water blends outside the US

Australia lists two gripe waters on its Register of Therapeutic Goods as listed medicines: Baby Gripe Water, with caraway, cardamom, cinnamon bark, clove bud and dill seed oils, and Hartley's Gripe Water, with coriander, dill seed and peppermint oils and chamomile. Neither register entry gives amounts.

No synergy involving dill has been demonstrated. A product page that cites the fennel colic trials to sell a dill product is borrowing evidence from a different plant.

Interactions & What NOT to Take

Photosensitizers. Methoxsalen used in PUVA therapy, plus any drug a pharmacist has flagged for sun sensitivity: additive phototoxicity is plausible from dill's furanocoumarins with heavy skin contact or concentrated topical use. Eating dill at food amounts is not the concern.

Mefenamic acid and other NSAIDs. Heidarifar's trial explicitly did not test dill alongside NSAIDs, so using both for cramps is untested rather than shown safe or additive.

Oxytocin and labor management. One labor trial compared dill seed decoction directly against oxytocin infusion. Dill is not something to add on one's own to an induced or augmented labor.

Glucose-lowering drugs. Named here only: the diabetes trials measured insulin and HOMA-IR, and that subject is covered in another issue. Monitor if taking dill capsules daily.

Sodium, propylene glycol and alcohol in infant products. The UK SmPC warns against combining its gripe water with other medicines containing propylene glycol or alcohol and flags its sodium load for infants on low-salt diets. Do not give two gripe waters, or a gripe water with simethicone or other colic drops, on the assumption that more products means more relief.

CYP450 and anticoagulants. No dill-specific human interaction data with named drugs turned up in this research. Those general topics are covered elsewhere in this batch.

Quality, Testing & Adulteration

Contamination in infant products sold under the gripe water name

• September 2007: FDA warned consumers not to use Baby's Bliss Gripe Water, apple flavor, code 26952V, expiration 10/08, after laboratory analysis confirmed Cryptosporidium in the product while investigating the illness of a six-week-old infant in Minnesota. About 17,600 bottles had been sold nationwide between November 2006 and September 2007. The distributor was MOM Enterprises of San Rafael, California, the company that does business today as Mommy's Bliss. The labeled ingredients were ginger and fennel extracts, water, glycerin, fructose, apple flavor, citric acid, bioflavonoid and grapefruit seed extracts. No dill.

• 2004: Sas, Enrione and Schwartz at Inova Fairfax Hospital for Children reported a 9-month-old girl in septic shock from Pseudomonas aeruginosa after ingestion of a contaminated gripe water, in The Pediatric Infectious Disease Journal.

• February 2019: Kingston Pharma recalled all lots of DG Baby Gripe Water, sold through Dollar General, for an undissolved citrus flavonoid ingredient, after a report of a one-week-old infant having difficulty swallowing it and three related complaints. That product's actives were also ginger and fennel.

• August 2023: FDA's warning letter to CalmCo over Colic Calm also pointed to a separate 26 July 2023 warning letter to Denison Pharmaceuticals citing current good manufacturing practice violations.

Little Remedies' own recall, announced 17 June 2025, involved its Honey Cough Syrup (Bacillus cereus), not the brand's gripe water.

Dill oil identity

The adulteration risks for dill seed oil are old and well characterized. The 1911 Codex flagged decarvonized oil by low specific gravity and Indian oil by its lower optical rotation and higher density. Today the equivalent is a gas chromatography profile: European seed oil should be carvone-dominant with limonene second and little or no dillapiole; herb oil should be phellandrene and dill ether dominant with no carvone; Indian seed oil shows dillapiole near a fifth of the total. A certificate that reports only "dill oil, conforms" without a carvone figure and a dillapiole figure has not addressed either substitution.

What third-party marks cover

A USP Verified or NSF mark on an herbal product means an auditor checked that the bottle contains what the panel declares, within contaminant limits, made under good manufacturing practice. It does not evaluate the age directions, the indication, or whether the herb named on the front is present at a meaningful amount, and no such mark makes a product suitable for a newborn.

Checks a parent or buyer can make from the box

• Is dill on the ingredient list at all? On the three US brands above, it is not.

• If it is, which oil, and how much per mL? The UK license states 2.3 mg terpeneless seed oil per 5 mL; a US product that lists "dill" with no amount has disclosed nothing measurable.

• Is there sodium bicarbonate, and how much sodium per dose? That figure matters more for an infant than the herb does.

• What is the sweetener: sucrose, fructose, agave, maltitol, glycerin? Each brings its own issue.

• For a homeopathic, the X number is a dilution: 3X means 1 part in 1,000. The listed herbs are present in trace amounts by design.

• Note the lot and expiration date: FDA's 2007 Cryptosporidium warning was issued by product code and expiration.

Special Considerations

Pregnancy and labor, where the Iranian obstetric trials sit. These are hospital studies, listed here so nobody mistakes them for home guidance. Hekmatzadeh and colleagues in Yasuj gave 10 g of dill seed boiled for 10 minutes in 100 mL of water, filtered, once at 3 to 4 cm of cervical dilatation, to the intervention group of a 153-participant trial against routine care in 2014, and reported shorter labor stages (except the second stage in first-time mothers) and better dilatation and effacement scores at 1 to 4 hours. Their 2019 follow-up randomized 100 women to the same decoction or 10 IU oxytocin in 1,000 mL Ringer solution and reported less severe anxiety after delivery in the dill group. Talebi and colleagues' 2020 systematic review pooled 2 randomized trials and reported shorter first (mean difference minus 43.66, 95 percent CI minus 52.99 to minus 34.33), second (minus 15.76) and third (minus 1.79) stages; the abstract does not state the unit. A substance trialed as a labor accelerant is a substance to avoid in medicinal amounts before labor.

Lactation. Carvone, dill seed's main compound, passes into breast milk: LactMed records it detected in the milk of 18 of 20 mothers given 30 mg of d-carvone, at an average 3.2 micrograms per liter, and notes that breastfed infants in the same research accepted carvone-flavored food more readily than formula-fed infants. Culinary dill while breastfeeding is ordinary food. Dill as a milk-supply remedy is untested on its own.

Infants and the one dill trial that exists. Sahib and colleagues in Baghdad randomized 28 infants aged 3 to 11 months with giardiasis to a boiled aqueous extract of dried dill leaf or metronidazole for 5 days, 14 per arm, publishing in 2014, and reported comparable parasite clearance. It had no placebo arm and reports no blinding, its authors wrote that subjective side effects could not be documented in infants that young, it was not about colic, and it is the only randomized infant dill trial this research found. It does not convert into a home dose.

How common infant gripe water use is. In the Avon Longitudinal Study of Parents and Children in southwest England, 13 percent of babies had been given gripe water before 8 weeks of age (Headley and Northstone 2007). A Sheffield survey of 200 mothers cited by Adhisivam found 64 percent gave it within the first month.

The history, as history. Adhisivam traces the formula to physicians treating infants for "fen fever," a malarial illness, in the 1840s. Widely repeated accounts, usually citing Blumenthal's 2000 history, date Woodward's product to 1851, made by a pharmacist with a Nottingham business, and the "Gripe Water" trademark to 1876, with the slogan "Granny told mother and mother told me." In 2012 Adhisivam quoted Indian gripe water advertising ("Granny gave mother and mother gave me," "Trusted by generations of mothers in India") as promotion that should be curbed.

Research Status & Evidence Quality

Tier 1, regulatory traditional-use approval: dill fruit for dyspeptic complaints, Commission E 1987. Not a trial result.

Tier 2, small randomized adult trials with a positive primary result: primary dysmenorrhea and vulvovaginal candidiasis, 70 analyzed and 60 randomized.

Tier 3, randomized adult evidence that conflicts: lipids, where the largest trial was null and the most positive was not truly randomized.

Tier 4, obstetric trials in Iran: labor duration and labor anxiety, single-center, open designs, hospital administration.

Tier 5, animal and in vitro: ileal spasmolysis, gastric antisecretory and mucosal protection, antimicrobial and antimycobacterial activity, estrous cycle effects in rats.

Tier 6, infant colic, for dill specifically: no controlled trial. A PubMed search on 19 October 2026 for Anethum graveolens in the title or abstract combined with infant, colic or children returned five records; none is a colic trial. The infant colic trials that do exist are fennel or fennel-blend trials (Alexandrovich 2003, 121 analyzed; Savino 2005, 88 analyzed). The observational infant gripe water data point the wrong way (Jain 2015, 335 mother-infant pairs, more colic, vomiting and constipation among users, cross-sectional).

Tier 7, lactation: no valid dill-only trial, per LactMed (revised May 2025).

Research Limitations

Nearly every human dill trial comes from Iran, most are single-center, several are single-blind or open, and preparations range from seed powder to leaf powder to hydroalcoholic aerial-part extracts to unstated tablets, so no two trials test the same material. None reported a carvone, limonene or dillapiole content for the product given. Several report within-group changes as if they were between-group effects; Mansouri's placebo arm saw the same triglyceride fall as the dill arm. Abstracts sometimes inflate their denominators: Haidari's abstract describes 42 patients as randomized when 48 were and 42 completed. No dose-ranging study exists for any indication, and no human pharmacokinetic study of a dill preparation was found.

Summary & Key Takeaways

Bottom Line

Dill is two crops with two unrelated oils: seed oil that is mostly carvone and limonene, herb oil that is mostly phellandrene and dill ether. Its adult case is a 1987 German traditional-use approval for dyspeptic complaints at 3 g of fruit a day, a small dysmenorrhea trial in which it matched a half-dose NSAID, and lipid results that two meta-analyses read differently. Its fame is gripe water, a 19th-century proprietary of alcohol, baking soda and dill oil that an earlier revision of an FDA import alert ordered detained at the border. No controlled trial found in this research has tested dill for infant colic.

Key Safety Points

• There is no evidence-based dill dose for an infant. A crying baby who is feeding poorly, vomiting, febrile or crying differently needs assessment, not a remedy.

• The UK's licensed dill gripe water is not for babies under one month, nor for infants with kidney problems, alkalosis or low calcium. US gripe waters have no FDA review of their directions.

• In infant products the excipients matter more than the herb: sodium bicarbonate, sugar or maltitol, propylene glycol, and historically alcohol, up to 9 percent in some brands.

• Infant gripe water products have been linked to Cryptosporidium contamination (2007), Pseudomonas septic shock (2004) and a recall involving a one-week-old (2019).

• Dill allergy is IgE mediated, has been reported as anaphylaxis, can be triggered by inhaled cooking vapor, and cross-reacts with other Apiaceae spices.

• Dill seed decoction is trialed as a labor accelerant. Avoid medicinal amounts in pregnancy.

• Know which oil you hold: European seed, dill weed, or Indian (Anethum sowa) with about 20 percent dillapiole.

Special Note

In an earlier revision of Import Alert 66-41, as archived, FDA recorded that its drug center "has safety concerns regarding the dill oil or dill water ingredients" in gripe water, and told every district to detain shipments so labeled. The alert does not explain the concern, no published FDA analysis of it turned up in the research for this issue, and the alert's current guidance no longer mentions gripe water. Meanwhile the American gripe water market reformulated around fennel and ginger, kept the old name, and moved into the dietary supplement category, where nobody reviews an infant direction before it is printed. The word survived; the dill mostly did not; and one of the few ingredients with controlled data on calming crying infants, plain sugar, was never the herb at all.

Researched and drafted with AI assistance. Every claim verified against primary sources and human-reviewed before publication. Last reviewed: October 19, 2026. Corrections: reply to any issue.

This newsletter is for education only and does not replace medical advice. Anyone taking medication, pregnant or breastfeeding, living with a diagnosed condition, or deciding what to give an infant should talk to a clinician before changing anything.