Magnesium is a cofactor in more than 300 enzyme systems, and the NIH Office of Dietary Supplements gives an adult male RDA of 400–420 mg a day, with a 350 mg daily upper limit for magnesium from supplements. Absorption studies favour soluble forms over oxide; the glycinate-specific evidence is two small trials pointing different ways; and trials in people with normal magnesium status show no consistent performance benefit.
The research divides cleanly. What is settled is magnesium’s roles and the intake figures built on them. What is not settled is which form absorbs best, and whether a supplement does anything for training in someone who is not short of the mineral. The table keeps the two apart, with the limit beside every finding.
What does magnesium do in the body?
The NIH Office of Dietary Supplements describes it as “a cofactor in more than 300 enzyme systems”, including those for protein synthesis and muscle and nerve function; it is also required for energy production, for bone and for making DNA and RNA. An adult body holds about 25 g, 50–60% of it in bone and most of the rest in soft tissue. Less than 1% sits in blood serum.
| Source | Design | Population and n | Dose and duration | What was measured | Finding, with its size | Limit or funding |
|---|---|---|---|---|---|---|
| NIH Office of Dietary Supplements, fact sheet, updated January 2026 | Agency summary of the evidence | The US population; intake data from national surveys | RDA 400 mg (men 19–30), 420 mg (men 31 and over), 310–320 mg (women); upper limit 350 mg a day from supplements and medicines, age 9 and over | Roles, requirements, intakes, the upper limit | 48% of Americans take in less magnesium from food and drink than their estimated average requirement | An intake figure, not a blood-test finding; serum holds under 1% of the body’s magnesium |
| Lindberg et al., J Am Coll Nutr, 1990 | Laboratory solubility measurements, then an oral load in volunteers | Normal volunteers | 25 mmol of magnesium as citrate or oxide, one load | Solubility; rise in urinary magnesium | Citrate 55% soluble in water, oxide virtually insoluble; oxide 43% soluble at simulated peak stomach acid; urinary rise larger after citrate (p < 0.05) | One dose; urinary magnesium is an indirect measure of absorption |
| Walker et al., Magnes Res, 2003 | Randomised, double-blind, placebo-controlled, parallel groups | 46 healthy people across four arms | 300 mg elemental magnesium a day as amino-acid chelate, citrate or oxide, or placebo, for 60 days | Serum, urinary, salivary and red-cell magnesium | Citrate gave the highest serum magnesium after one dose (P = 0.026) and at 60 days (P = 0.006); citrate and chelate raised 24-hour urinary magnesium more than oxide at 60 days (P = 0.033); oxide no different from placebo | Small arms; the chelate’s amino acid is not named in the abstract |
| Schuette et al., JPEN, 1994 | Randomised, double-blind crossover with a labelled magnesium tracer | 12 patients with ileal resection | 100 mg of labelled magnesium as diglycinate or oxide | Fractional absorption | 23.5% against 22.8% overall, no difference; 23.5% against 11.8% (p < 0.05) in the 4 patients with the most impaired absorption; diglycinate better tolerated by all patients | A clinical population, not people who train; 12 people; the advantage sits in a subgroup of four |
| Pajuelo et al., Nutrients, 2024 | Randomised, double-blind crossover | 40 healthy adults aged 20–55 | 375 mg elemental magnesium as oxide, citrate, bisglycinate or a microencapsulated oxide; single doses after a 7-day low-magnesium diet | Plasma magnesium at 1, 4 and 6 hours | Oxide raised it at 1 hour (+7.1%), citrate at 4 hours (+7.5%), bisglycinate at no time point; areas under the curve did not differ between products | Funded by the maker of the microencapsulated product, which supplied all four; two authors were its employees; six hours of plasma is not status |
| Zhang et al., Nutrients, 2017 | Narrative review | Animal and human studies | Varied | Exercise performance; magnesium requirements | The need for magnesium rose with physical activity; human performance findings were inconsistent | Not a systematic review; the authors call for larger, more rigorous trials |
| Bomar et al., Nutrients, 2025 | Randomised, double-blind, placebo-controlled crossover | 15 regular exercisers aged 18–40 with normal magnesium status | 300 mg magnesium chloride twice a day for 9 days | VO2max, 30-second sprint power, 10 km time trial | VO2max 41.3 against 44.4 mL/kg/min (p = 0.005); sprint mean power 415 against 439 W (p = 0.03); time trial unchanged (p = 0.89) | 15 people, 9 days, a different salt from glycinate; company-funded, with no stated sponsor role |
What is established
Magnesium’s roles, and the numbers built on them. The Office of Dietary Supplements gives the RDA as 400 mg a day for men aged 19–30 and 420 mg from 31, counting food, drink, supplements and medicines together, and the upper limit as 350 mg a day for magnesium from supplements and medicines only, for everyone aged 9 and over. The limit is set on the gut: high supplemental doses often cause diarrhoea. That is why the upper limit can sit below the requirement without contradicting it. The magnesium in food does not count against the 350 mg; the magnesium in a capsule does.
And the direction of the absorption data. Forms that dissolve well in liquid are generally absorbed better than other forms, in the Office’s summary, and oxide is the documented laggard: virtually insoluble in water in Lindberg and colleagues’ measurements, and no different from placebo over 60 days in Walker and colleagues’ trial. The Office names aspartate, citrate, lactate and chloride as the better-absorbed forms; glycinate is not on that list. The arithmetic of the upper limit is worked through on can you take too much magnesium and the daily figures on how much magnesium per day.
What do magnesium absorption studies show?
That solubility predicts absorption and oxide trails. Lindberg and colleagues measured citrate at 55% soluble in water against virtually insoluble oxide; Walker and colleagues found citrate gave the highest serum magnesium over 60 days and oxide no different from placebo. Glycinate is thinner: equal to oxide overall in 12 patients with ileal resection (Schuette, 1994), and the only form not to raise plasma magnesium in a 2024 crossover of 40 adults.
Why are magnesium absorption studies hard to compare?
Because each one measures a different thing. Lindberg (1990) used urinary magnesium after a single load; Walker (2003) used serum, urine and saliva over 60 days; Schuette (1994) used a labelled tracer in patients with ileal resection; Pajuelo (2024) used finger-prick plasma for six hours. Doses ran from a 100 mg tracer to a 25 mmol load, and serum holds under 1% of the body’s magnesium. The direction can be read across them; a ratio cannot.
That is also why the 2024 crossover is printed here even though it is unwelcome for the form TARE sells. It was funded by the maker of a competing ingredient, and a six-hour plasma curve is not magnesium status — but bisglycinate was the one form of four that did not significantly raise plasma magnesium, and leaving it out would be the selective citation this Journal exists to refuse. Its areas under the curve did not differ between products either, which cuts against every headline the trial could be used for, including the sponsor’s.
What does the research on magnesium and exercise show?
No consistent benefit for people who are not short of magnesium. Zhang and colleagues’ 2017 review reports that the need for magnesium rises with physical activity, but found human performance findings inconsistent. In a 2025 crossover, 15 regular exercisers with normal magnesium status took 300 mg of magnesium chloride twice a day for nine days: VO2max was 41.3 mL/kg/min after magnesium against 44.4 after placebo, and sprint power was lower too. Fifteen people, nine days, a different salt.
The detail of that trial, and the other questions a lifter asks before taking magnesium every day, are on magnesium glycinate side effects. Its authors’ own recommendation is that regular exercisers without low blood magnesium should not supplement their diet with magnesium, and a page selling magnesium should be able to print that sentence.
How strong is the evidence for magnesium glycinate?
Thin, and it points both ways. The older trial, Schuette and colleagues in 1994, found diglycinate and oxide absorbed about equally across 12 patients with ileal resection, 23.5% against 22.8%, with glycinate ahead only in the four whose absorption was most impaired. A 2024 crossover found bisglycinate the only one of four forms not to raise plasma magnesium, in a trial funded by a competing ingredient’s maker. What remains is tolerability, not an absorption multiple.
What the research does not show
- An absorption multiple for glycinate. No study here shows glycinate absorbed twice as well as oxide in people who train. The doubling in the 1994 trial belongs to four patients with ileal resection.
- Status from a six-hour curve. Serum holds under 1% of the body’s magnesium, so a plasma curve after one dose answers a narrower question than whether a form works over weeks.
- A performance effect in people with normal magnesium status. The 2017 review found the human findings inconsistent, and the 2025 crossover moved two measures the wrong way.
- A long head-to-head. The longest comparison of forms here ran 60 days, in 46 people.
- A ratio anyone should quote. The studies use different measures, doses and populations; the direction is consistent, the precision is not.
Where Night Work sits
Night Work is magnesium glycinate in capsules: 275 mg of elemental magnesium from 2,500 mg of magnesium glycinate per three-capsule serving (label figures), 30 servings. It is the only supplemental magnesium in the TARE range — 275 mg, 75 mg under the 350 mg tolerable upper intake level for supplemental magnesium, which is room a multivitamin or an antacid can use without saying so. The case for glycinate is tolerability — in the 1994 trial diglycinate was better tolerated by all 12 patients — and oxide’s weak showing, not an absorption multiple. The certificates and what they do and do not cover are on Testing & Certification.
The long-form pages: magnesium glycinate vs oxide, magnesium glycinate vs citrate, magnesium bisglycinate vs glycinate, elemental magnesium, how to read a label and how to read supplement research. The same evidence map for the other two ingredients: creatine research summary and collagen research summary. Every form compared side by side: supplement forms compared.
Sources
- NIH Office of Dietary Supplements. Magnesium: Fact Sheet for Health Professionals (updated 6 January 2026). NIH Office of Dietary Supplements
- Lindberg JS, Zobitz MM, Poindexter JR, Pak CY. J Am Coll Nutr 1990;9(1):48-55. PubMed 2407766
- Walker AF, Marakis G, Christie S, Byng M. Magnes Res 2003;16(3):183-191. PubMed 14596323
- Schuette SA, Lashner BA, Janghorbani M. JPEN J Parenter Enteral Nutr 1994;18(5):430-435. PubMed 7815675
- Pajuelo D, Meissner JM, Negra T, et al. Nutrients 2024;16(24):4367. PubMed 39770988
- Zhang Y, Xun P, Wang R, Mao L, He K. Nutrients 2017;9(9):946. PubMed 28846654
- Bomar MC, Ewell TR, Brown RL, et al. Nutrients 2025;17(5):915. PubMed 40077784