Watermelon: Food Safety, FODMAPs, and Blood Sugar
Watermelon is a healthy food that causes three specific, entirely avoidable problems, and all three are widely misunderstood. The first is that melon is a repeat vehicle for foodborne illness — not because the flesh is contaminated, but because the knife carries bacteria from the outside of the rind straight into it, and cut melon left warm is close to an ideal growth medium. The second is that watermelon is high in more than one FODMAP group, which makes it one of the more common hidden triggers for people with irritable bowel syndrome who have already given up onions and wheat and cannot work out what is left. The third is the glycaemic index question, where the honest answer — high index, low load — is routinely mangled in both directions, scaring off people with diabetes for whom a normal serving is a low glycaemic load, and reassuring people eating half a melon, which is a real carbohydrate load. None of this makes watermelon a food to give up. All of it is worth knowing.
Table of Contents
- Why Melon Specifically
- The Outbreak Record, With Real Numbers
- The Knife Is the Mechanism
- Cut Melon Grows Bacteria Fast
- Pre-Cut Melon From the Store
- Safe Handling, Step by Step
- Who the Research Raises Concerns For
- FODMAPs: Why Watermelon Sets Off Irritable Bowel Syndrome
- Fructose Malabsorption in Plain Language
- Glycaemic Index Versus Glycaemic Load
- What This Means If You Have Diabetes
- Other Cautions Worth a Sentence
- Key Research Papers
- Connections
- Featured Videos
Why Melon Specifically
Melons carry an unusual combination of properties that together make them a better vehicle for bacteria than almost any other fruit:
- They grow lying on soil. The fruit rests directly on the ground for weeks, in contact with irrigation water, soil, animal droppings and whatever runoff reaches the field. Apples hang in the air; melons do not.
- The flesh is not acidic. Watermelon flesh has a pH around 5.2–5.6 — nearly neutral. Most fruits are protected by acidity: an orange or a strawberry sits near pH 3.5, low enough to suppress Salmonella and Listeria. Melon flesh offers no such protection.
- It is wet, sugary and nutrient-rich — a laboratory would struggle to design a friendlier medium.
- Nobody cooks it. There is no kill step anywhere between the field and the plate.
- It is cut with a knife that goes through the outside first. This is the whole story, and it gets its own section below.
Cantaloupe is worse than watermelon on the first point — its netted, corky rind is far harder to clean than watermelon's smooth skin, and organisms lodge in the netting where washing cannot reach. But watermelon is not exempt, and the outbreak record proves it.
The Outbreak Record, With Real Numbers
Researchers at the Centers for Disease Control and Prevention reviewed every outbreak reported to the national Foodborne Disease Outbreak Surveillance System between 1973 and 2011 in which the implicated food was a single type of melon. The findings:
- 34 outbreaks, causing 3,602 illnesses, 322 hospitalisations, 46 deaths and 3 fetal losses.
- Cantaloupe accounted for 19 of the outbreaks (56%), watermelon for 13 (38%), honeydew for 2 (6%).
- Salmonella was the most common cause (19 outbreaks, 56%), followed by norovirus (5 outbreaks, 15%).
- The rate rose over time — from 0.5 outbreaks a year in 1973–1991 to 1.3 a year in 1992–2011.
- Where the point of contamination could be determined, it happened most often during growth, harvest, processing or packaging (13 of 20 outbreaks, 65%) rather than in the kitchen.
- Among 13 outbreaks with information on where the melons came from, melons imported from Mexico and Central America were implicated in 9 (69%) and domestically grown melons in 4 (31%).
Watermelon accounting for well over a third of melon outbreaks contradicts any idea that the smooth rind makes it safe.
A newer review led by the Centers for Disease Control and Prevention (CDC) covered 2012–2021 and counted 18 outbreaks linked to melons in the United States, causing 821 illnesses, 251 hospitalisations and 10 deaths. More than half of the melon-linked Salmonella outbreaks in that period were linked to cantaloupe. Where the growing area was known, most of the melons were produced domestically (4 of 7, 57%), and traceback and epidemiologic investigations showed that contamination can occur at every point from farm to fork.
The single most instructive melon outbreak is a cantaloupe one. In 2011, Listeria monocytogenes traced to whole cantaloupes from one Colorado farm caused 147 confirmed cases across 28 states. Of 145 patients with hospitalisation data, 143 — 99% — were hospitalised. Thirty-three died. Eighty-six percent of patients were 60 or older; seven infections occurred in pregnant women and newborns, with one related miscarriage. Investigators found unsanitary conditions in the farm's packing facility. Raw produce had rarely been recognised as a listeriosis vehicle before; after 2011, nobody made that mistake again.
The Knife Is the Mechanism
Here is the part that changes behaviour once it is understood. The inside of an intact melon is essentially sterile. The problem is transport, and the vehicle is the knife.
The rind carries whatever the field and the supply chain put on it — soil organisms, irrigation-water contaminants, handling from a dozen people, contact with other produce in a crate. When the blade passes through that rind it picks the organisms up and drags them across every square millimetre of flesh it then cuts. A melon that was safe as long as it stayed whole becomes contaminated throughout in a single stroke.
Researchers demonstrated this directly. Working with cantaloupes deliberately inoculated with Listeria monocytogenes on the rind surface, they measured the transfer of the organism from rind to fresh-cut pieces during cutting, and tested how well two sanitiser washes reduced it. Listeria survived on the rinds for up to 15 days at 4 and 20 °C, although populations declined by about 1 to 2 log units, and fresh-cut pieces prepared from inoculated melons stored at 4 °C for 24 hours were positive for Listeria. Washing the whole melons for 2 minutes in a solution of 1,000 ppm chlorine or 5% hydrogen peroxide gave a 2.0- to 3.5-log reduction on the melon surface (each log is a tenfold drop), and fresh-cut pieces prepared from the sanitised melons were negative. That study tested chlorine and hydrogen-peroxide washes, not plain water. For home kitchens, the US Food and Drug Administration (FDA) produce guidance states that even when the skin will not be eaten, produce is washed first so that dirt and bacteria are not transferred from the surface when it is cut, and that firm produce such as melons is scrubbed with a clean produce brush — an instruction that sounds absurd until it is clear what it is for.
Two consequences follow, and neither is intuitive:
- The outside is washed even though nobody eats the outside. The washing is not for the rind's sake.
- The washing comes before cutting, not after. Rinsing cut flesh does not undo the transfer, it just makes the melon wet.
Cut Melon Grows Bacteria Fast
Contamination alone rarely makes a healthy person ill — a handful of organisms usually will not. What makes people ill is growth, and this is where melon's near-neutral pH stops being trivia and starts being the problem.
The work has been done. A study of Salmonella growth in cantaloupe, watermelon and honeydew melon flesh established that these fruits support the organism's multiplication at warm temperatures. Later work measured the growth of Salmonella Enteritidis, and separately of Listeria monocytogenes, in melon, watermelon and papaya pulps held at different temperatures for different lengths of time — a design chosen precisely because the practical question is not "can it grow" but "how fast, and at what temperature."
The measured results, by temperature:
- At room temperature and above, populations can rise from harmless to infectious within a few hours. Salmonella growth on melon pieces was "rapid and prolific" at 23 °C over 24 hours, and Listeria in watermelon pulp (pH 5.50) doubled every 2.17 hours at 20 °C and every 1.00 hour at 30 °C.
- In the cold, Salmonella on melon pieces did not increase over 24 hours at 5 °C — but little or no decrease was seen either. Listeria placed directly on fresh-cut cantaloupe survived, but did not grow, during 15 days at 4 °C; growth was evident within 4 hours at 8 and 20 °C.
- A cool fridge is not enough to stop Listeria — in watermelon pulp at 10 °C its growth was slowed but not stopped, with a doubling time of 13.03 hours. Refrigerating cut melon is a large improvement, not a permanent solution, which is why cut melon has a shelf life measured in days rather than weeks.
The FDA's guidance for eating outdoors gives the standard rule for perishable food: not more than 2 hours in the "Danger Zone" between 40 °F and 140 °F, or 1 hour if the outdoor temperature is above 90 °F. The fruit platter at a summer barbecue that sat out all afternoon is the single most common way this goes wrong in a home, and it is the reason melon shows up in outbreak reports from picnics, buffets and catered events out of all proportion to how much of it people eat.
Pre-Cut Melon From the Store
Pre-cut melon in a plastic tub is convenient and it is also the highest-risk form of the fruit, for reasons that follow directly from everything above:
- It has already been through the cutting step, on industrial equipment processing many melons, so one contaminated fruit can seed an entire production run.
- It has been in transit and on display for an unknown time at an unknown temperature.
- It is often mixed with other cut fruit, widening the exposure.
- You cannot inspect it, and there is no cooking step to fall back on.
Pre-cut melon has been the identified vehicle in multistate Salmonella outbreaks in the United States in recent years, with recalls spanning several states and multiple retail chains at a time. The pattern repeats often enough that it is a known hazard of the product rather than a run of bad luck.
A whole melon cut at home skips the industrial cutting step, and it is cheaper per kilogram as well. For pre-cut produce such as half of a watermelon, the FDA's produce guidance is to choose only items that are refrigerated or surrounded by ice, and to refrigerate all produce bought pre-cut or packaged.
Safe Handling, Step by Step
- Whole melons. An intact melon protects its own flesh.
- Whole melons keep at cool room temperature until they are cut. Long refrigeration of an uncut melon causes chilling injury and degrades flavour and pigment.
- Scrubbing the rind under running water with a clean produce brush, just before cutting, is the FDA's guidance for firm produce such as melons. The same guidance states that washing fruits and vegetables with soap, detergent or commercial produce wash is not recommended, because produce is porous and can absorb them. The rind-transfer study above tested chlorine and hydrogen-peroxide washes, not plain-water scrubbing.
- Drying the melon with a clean cloth or paper towel further reduces bacteria on the surface, according to the FDA page, and stops the wet rind sliding under the knife.
- A clean board and a clean knife — not the ones that just did raw chicken. The FDA page lists washing cutting boards, utensils and countertops with soap and hot water between raw meat, poultry or seafood and produce that will not be cooked.
- Cut fruit into the fridge within two hours, or within one hour above 90 °F — the FDA's outdoor-eating limit for perishable food — and kept covered.
- Cut melon keeps three to four days. Melon that has gone slimy, fizzy, sour-smelling or watery at the bottom of the container is spoiled; fermentation is a sign that something has been multiplying.
- Kept cold at gatherings — on ice, or brought out in batches, rather than one large platter that sits in the sun.
Who the Research Raises Concerns For
For a healthy adult, a bout of Salmonella from melon means several unpleasant days. For some people it is far more serious, and those groups are exactly the ones the 2011 listeriosis outbreak hit:
- Adults over 60. Eighty-six percent of the cases in that outbreak, and almost all of the deaths.
- Pregnant women. Listeria crosses the placenta and causes miscarriage, stillbirth and neonatal infection even when the mother has only mild symptoms. Pregnancy guidance in most countries specifically warns against pre-cut melon and unrefrigerated fruit platters, and this is why.
- Anyone immunocompromised — cancer treatment, transplant medication, advanced human immunodeficiency virus (HIV), long-term corticosteroids, dialysis.
- Infants and young children, whose infectious dose is lower.
For these groups the risk described on this page lies in how melon is handled rather than in the fruit itself: a whole melon, scrubbed, cut at home, refrigerated promptly and eaten within days, on one side; the pre-cut tub and the room-temperature buffet platter, the highest-risk forms, on the other.
FODMAPs: Why Watermelon Sets Off Irritable Bowel Syndrome
FODMAP stands for Fermentable Oligosaccharides, Disaccharides, Monosaccharides And Polyols — a family of short-chain carbohydrates that share two properties: they are poorly absorbed in the small intestine, and they are rapidly fermented by bacteria in the colon. Being poorly absorbed, they draw water into the small bowel osmotically; being rapidly fermented, they generate gas in the large bowel. In a person with a normal gut, that produces nothing noticeable. In someone with the visceral hypersensitivity that characterises irritable bowel syndrome, it produces pain, distension, urgency and bloating.
The approach was formalised by a research group at Monash University, and the trial evidence is strong for a dietary intervention. A randomised controlled feeding study found that a diet low in FODMAPs reduced overall gastrointestinal symptom scores in people with irritable bowel syndrome (IBS) compared with a typical Australian diet, with the improvement seen across bloating, pain and wind. The underlying measurement work — quantifying these short-chain carbohydrates in common fruits and vegetables by high-performance liquid chromatography — is what made the food lists possible in the first place.
Watermelon's problem is that it is high in several FODMAP categories at once. Most fruits that cause trouble do so through one route. Watermelon contributes excess fructose (fructose in excess of glucose), fructans, and the polyol mannitol. That triple loading is why the Monash program classifies watermelon as a high-FODMAP food at ordinary serving sizes, and why it is one of the fruits most often named when someone with irritable bowel syndrome finally identifies their trigger.
Three things make watermelon particularly easy to miss as a culprit:
- Portions are enormous. Nobody eats 400 g of apple. Plenty of people eat 400 g of watermelon without thinking about it, and FODMAP effects are strictly dose-dependent.
- It reads as innocent. It is water and fruit — the last thing anyone suspects when the obvious offenders have already been removed.
- The delay. Symptoms follow fermentation in the colon, so they can arrive hours after eating, which breaks the mental link to the meal.
For people with irritable bowel syndrome who have not tested watermelon, a deliberate trial is one way to find out: removing it completely for two weeks, then reintroducing a measured portion on a day when nothing else is unusual. A small portion may be tolerated even when a large one is not — the low-FODMAP approach is about dose and threshold, not permanent elimination. See also the site's page on irritable bowel syndrome.
Fructose Malabsorption in Plain Language
The fructose part of the story is worth explaining on its own, because it is the mechanism behind a common and frequently undiagnosed complaint.
Fructose is absorbed across the small intestine by a transporter called GLUT5, which has limited capacity. There is a second route — GLUT2 — which is much more efficient but effectively requires glucose to be present alongside the fructose. This produces a rule of thumb that explains most of the food lists: a fruit with roughly equal glucose and fructose is usually well tolerated, while a fruit with fructose in excess of glucose leaves the surplus fructose stranded with no efficient way across, so it passes into the colon and is fermented.
Watermelon is an excess-fructose fruit, which is why it appears alongside apples, pears, mangoes and honey on malabsorption lists, while bananas, oranges and berries generally do not. Reviews of dietary fructose intolerance describe both the mechanism and the substantial overlap between fructose malabsorption and irritable bowel syndrome symptoms; work on fructose taken with and without simultaneous glucose confirms that the co-ingestion of glucose materially improves tolerance.
The practical trick that follows: eating a small portion of watermelon as part of a mixed meal is better tolerated than eating a large amount alone. Other food slows gastric emptying, spreads the fructose load over time, and supplies glucose alongside it. Half a cup after lunch is a very different exposure from a plate of watermelon on an empty stomach on a hot afternoon.
Glycaemic Index Versus Glycaemic Load
This is the watermelon question that is misunderstood most often, and the distinction is genuinely useful once it clicks.
Glycaemic index measures how fast the carbohydrate in a food raises blood glucose, compared with pure glucose at 100. Crucially, it is measured on a fixed 50 g dose of available carbohydrate, not on a normal portion. Watermelon's index in the international reference tables is high — in the low-to-mid 70s in commonly published values, with individual test results ranging higher. That is comparable to white bread, and it is where "watermelon spikes your blood sugar" comes from.
Glycaemic load asks the question that actually matters: given how much carbohydrate is in a real serving, how much does eating it move your blood glucose? The formula is the index multiplied by the grams of available carbohydrate in the serving, divided by 100.
Do the arithmetic for watermelon. U.S. Department of Agriculture (USDA) data gives 7.55 g of total carbohydrate and 0.4 g of fibre per 100 g, so about 7.15 g of available carbohydrate per 100 g. A standard 120 g reference serving therefore contains about 8.6 g. At a glycaemic index of around 72:
- Glycaemic load ≈ 72 × 8.6 / 100 ≈ 6 — comfortably in the "low" band, which is usually defined as 10 or under.
Now see why the index alone is so misleading. To get the 50 g of carbohydrate that a glycaemic index test uses, you would have to eat about 700 g of watermelon — roughly five cups. The high index describes the behaviour of an amount of watermelon nobody eats in one sitting. That is the entire trick, and it applies to other watery, high-index foods too.
A 2018 review of 73 studies looked at the concept more broadly and found the relationship between glycaemic index and disease outcomes equivocal; it concluded that other measures of diet quality, such as fibre or whole grains, may be more likely to predict health outcomes. Watermelon remains a textbook example of index and load pointing in opposite directions for the same food.
The one honest caveat: watermelon has essentially no fibre and no fat to slow absorption, so what carbohydrate there is arrives quickly. A modest portion is a genuinely low-impact food. An enormous portion — a 700 g wedge is 43 g of sugar — is a real carbohydrate load arriving fast, and calling that "low glycaemic load" because the reference serving is small would be dishonest. Portion is the whole argument in both directions.
One small trial measured blended watermelon directly. In a crossover study of 20 adolescents with overweight or obesity, 240 mL of blended watermelon (flesh and rind) gave significantly lower blood glucose at 20 and 40 minutes (P < 0.01) than a sugar-sweetened drink with the same calories, and the rise in hunger and desire to eat came later.
What This Means If You Have Diabetes
- Watermelon is not off the list. The high index is an artefact of the testing method, not a description of a normal serving.
- Portion is the control. One to one-and-a-half cups (150–225 g) is roughly 11–16 g of available carbohydrate — about the same as a small apple, and easy to count.
- With something else. Protein, fat or fibre alongside — nuts, cheese, yoghurt, or simply as part of a meal — blunts the rise substantially.
- Whole fruit versus juice. Juicing removes the volume that limits how much is consumed; a glass of watermelon juice represents far more fruit than would have been chewed.
- Individual responses differ. Glycaemic responses vary a great deal between individuals. With a glucose meter or a continuous monitor, a measured portion checked at 30, 60 and 120 minutes shows more about one person's response than any table can.
Other Cautions Worth a Sentence
Potassium restriction. Watermelon is low in potassium per 100 g but is eaten in very large portions, which is exactly how people on renal potassium restriction get caught. In a potassium count, the portion matters more than the per-100 g figure.
Allergy. Watermelon is a common trigger of oral allergy syndrome in people allergic to ragweed pollen — itching or tingling of the lips, mouth and throat within minutes of eating, caused by cross-reactivity between pollen proteins and similar proteins in the fruit. It is usually mild and often improves when the fruit is cooked (which denatures the proteins), though cooked watermelon is not much of a dish. True systemic watermelon allergy exists but is uncommon.
Very large quantities. Eating an entire melon in a sitting delivers a large osmotic and fructose load and predictably causes loose stools and cramping in almost anyone, irritable bowel syndrome or not. This is physiology, not intolerance.
The seeds are fine. Swallowing black watermelon seeds is harmless — nothing grows in your stomach. Roasted and salted, they are a traditional snack across the Middle East, Africa and Asia, and are a genuine source of protein, magnesium and fat, unlike the flesh. Analytical work on watermelon seed composition and seed oil confirms a nutrient profile quite unlike that of the fruit around it.
The rind is edible and traditionally pickled or stir-fried — but only after the same scrubbing the safety section describes, since the rind is the part that carried the field with it.
Key Research Papers
- Walsh KA, Bennett SD, Mahovic M, Gould LH. Outbreaks associated with cantaloupe, watermelon, and honeydew in the United States, 1973-2011. Foodborne Pathogens and Disease. 2014;11(12):945-952. — doi:10.1089/fpd.2014.1812
- McCollum JT, Cronquist AB, Silk BJ, et al. Multistate outbreak of listeriosis associated with cantaloupe. New England Journal of Medicine. 2013;369(10):944-953. — doi:10.1056/NEJMoa1215837
- Ukuku DO, Fett W. Behavior of Listeria monocytogenes inoculated on cantaloupe surfaces and efficacy of washing treatments to reduce transfer from rind to fresh-cut pieces. Journal of Food Protection. 2002;65(6):924-930. — doi:10.4315/0362-028X-65.6.924
- Golden DA, Rhodehamel E, Kautter DA. Growth of Salmonella spp. in cantaloupe, watermelon, and honeydew melons. Journal of Food Protection. 1993;56(3):194-196. — doi:10.4315/0362-028X-56.3.194
- Penteado AL, Leitão MF. Growth of Salmonella Enteritidis in melon, watermelon and papaya pulp stored at different times and temperatures. Food Control. 2004;15(5):369-373. — doi:10.1016/s0956-7135(03)00099-9
- Penteado AL, Leitão MF. Growth of Listeria monocytogenes in melon, watermelon and papaya pulps. International Journal of Food Microbiology. 2004;92(1):89-94. — doi:10.1016/j.ijfoodmicro.2003.08.020
- Bowen A, Fry A, Richards G, Beuchat L. Infections associated with cantaloupe consumption: a public health concern. Epidemiology and Infection. 2006;134(4):675-685. — doi:10.1017/s0950268805005480
- Halmos EP, Power VA, Shepherd SJ, Gibson PR, Muir JG. A diet low in FODMAPs reduces symptoms of irritable bowel syndrome. Gastroenterology. 2014;146(1):67-75. — doi:10.1053/j.gastro.2013.09.046
- Gibson PR, Shepherd SJ. Evidence-based dietary management of functional gastrointestinal symptoms: the FODMAP approach. Journal of Gastroenterology and Hepatology. 2010;25(2):252-258. — doi:10.1111/j.1440-1746.2009.06149.x
- Muir JG, Rose R, Rosella O, et al. Measurement of short-chain carbohydrates in common Australian vegetables and fruits by high-performance liquid chromatography (HPLC). Journal of Agricultural and Food Chemistry. 2009;57(2):554-565. — doi:10.1021/jf802700e
- Fedewa A, Rao SSC. Dietary fructose intolerance, fructan intolerance and FODMAPs. Current Gastroenterology Reports. 2014;16(1):370. — doi:10.1007/s11894-013-0370-0
- Latulippe ME, Skoog SM. Fructose malabsorption and intolerance: effects of fructose with and without simultaneous glucose ingestion. Critical Reviews in Food Science and Nutrition. 2011;51(7):583-592. — doi:10.1080/10408398.2011.566646
- Atkinson FS, Brand-Miller JC, Foster-Powell K, Buyken AE, Goletzke J. International tables of glycemic index and glycemic load values 2021: a systematic review. The American Journal of Clinical Nutrition. 2021;114(5):1625-1632. — doi:10.1093/ajcn/nqab233
- Atkinson FS, Foster-Powell K, Brand-Miller JC. International tables of glycemic index and glycemic load values: 2008. Diabetes Care. 2008;31(12):2281-2283. — doi:10.2337/dc08-1239
- Canning M, Bazaco M, Weed B, et al. Outbreaks linked to melons, United States, 2012-2021. Journal of Food Protection. 2025;88(7):100536. — doi:10.1016/j.jfp.2025.100536
- Rasmussen C, Rosas M, Gallardo I, et al. Effects of blenderized watermelon consumption on satiety and postprandial glucose in overweight and obese adolescents. Metabolism Open. 2025;25:100345. — doi:10.1016/j.metop.2025.100345
- Vega-López S, Venn BJ, Slavin JL. Relevance of the glycemic index and glycemic load for body weight, diabetes, and cardiovascular disease. Nutrients. 2018;10(10):1361. — doi:10.3390/nu10101361
- Eke R, Ejiofor E, Oyedemi S, Onoja S, Omeh N. Evaluation of nutritional composition of Citrullus lanatus Linn. (watermelon) seed and biochemical assessment of the seed oil in rats. Journal of Food Biochemistry. 2021;45(6):e13763. — doi:10.1111/jfbc.13763
- Petkowicz C, Vriesmann L, Williams P. Pectins from food waste: extraction, characterization and properties of watermelon rind pectin. Food Hydrocolloids. 2017;65:57-67. — doi:10.1016/j.foodhyd.2016.10.040
- US Food and Drug Administration — Selecting and Serving Produce Safely (the official wash-before-cutting and scrub-firm-produce guidance).
- US Food and Drug Administration — Handling Food Safely While Eating Outdoors (the 2-hour, or 1-hour above 90 °F, limit for perishable food).
- Monash University FODMAP program — About FODMAPs and IBS, the group that developed and maintains the food classification.
- Live search: PubMed: pre-cut melon and Salmonella outbreaks
- Live search: PubMed: watermelon glycaemic index and load
Connections
- Watermelon — main topic page
- Watermelon — Benefits Deep Dive
- Watermelon: History and Origins
- Watermelon, Citrulline, and Blood Flow
- Watermelon Lycopene and Heart Health
- Watermelon for Hydration and Electrolytes
- Irritable Bowel Syndrome (IBS)
- Small Intestinal Bacterial Overgrowth (SIBO)
- Salmonella
- Listeria monocytogenes
- Bacteria — all pathogens
- Cantaloupe
- Cucumber
- Potassium
- All Food topics
- Gastroenterology