Three of the Eight Numbers on the Back of This Bottle Match Their Trials

Memoractiv prints eight lines with a milligram figure beside each one. I set every figure next to the trial that used it, counted the caffeine in the capsule, and read the return terms.

Harlan Voss

Staff Writer, The Sunday DeskSeptember 20, 2026

This is a paid advertisement. Sponsored by Thorne Research, Inc.

Harlan Voss is the pen name of this desk.

Everything on this page about the bottle comes from the maker's own product listing, the panel printed on it and its published return terms, all read on September 20, 2026.

The four customer reviews quoted below are real and carry the store's own purchase mark; the small portraits beside them are illustrations, not the reviewers.

Most Sunday mornings I read a label the way other people read box scores.

This one took longer than usual, and not because it was badly written.

It took longer because the numbers are mostly real, and real numbers are slower to check than vague ones.

A panel that says "proprietary blend, 1,470 mg" can be read in four seconds, because there is nothing in it to read.

This panel prints eight lines, and every one of them has a milligram figure beside it.

That's rarer on this desk than you'd think, and it's the reason this bottle got a full morning instead of ten minutes.

So let me do the thing this bottle makes possible, before anything else.

Here is the panel, as printed, in the order it prints, for one serving of two capsules.

Read it the way you would read a scorecard.

Acetyl-L-carnitine — 250 milligrams.

Ashwagandha root extract, KSM-66 — 300 milligrams.

Bacopa extract, 20 per cent bacosides — 100 milligrams.

Caffeine — 50 milligrams.

Ginkgo phytosome — 240 milligrams.

Lutein — 10 milligrams. Zeaxanthin — 2 milligrams.

Trans-pterostilbene — 66 milligrams.

Eight lines, eight numbers, two capsules, and not one blend among them.

Eight lines, eight numbers. Three of them land where their trial landed.

Now the scoring, which is the whole of this page.

Three of those eight numbers are the number their own trial used; one is a third of its own; one counts a complex rather than the plant inside it; two have no trial I can hand you.

Memoractiv is made by Thorne, a supplement company in Summerville, South Carolina, with a street address, a phone number that answers on weekdays and its own manufacturing.

It comes as sixty capsules in a white bottle, taken two at a time.

By the maker's own rating, it is the best-received bottle I have looked at: 4.6 from 79 buyers, each review carrying the store's own purchase mark.

That is a real signal about satisfaction and it is not a signal about anything else, which is a distinction this page will keep making.

A high score tells you people liked what arrived. It does not tell you what any line on the panel did.

So here is the morning's work, in order.

The bottle and the rule about bedtime; then a short lesson in how a panel like this one is read at all.

Then the score and the four people who wrote in, then the eight lines, one at a time.

Each one beside the trial that tested that ingredient, and at what dose.

Then the words the panel uses without explaining: standardized, phytosome, co-crystal.

Then the arithmetic of one serving against two, the whole of the small print, and what the eight trials have in common.

What I found is three lines at trial dose, one at a third of it, one complex that doesn't say how much plant is inside it, two with no trial I can hand you, 50 milligrams of caffeine in a capsule, and sixty days to send it back minus a quarter.

Sixty Capsules, Two to a Serving, Thirty Servings, and a Line About Bedtime

This is what arrives.

The length matters more than it sounds.

Two capsules almost an inch long, swallowed together, is the whole serving, and some people will find that a lot to take at once.

I timed myself on the first morning out of curiosity, with a full glass of water.

Two went down without trouble; I would not want to try it with a sip.

The direction is short.

Take two capsules one to two times a day, away from bedtime, or as your own clinician advises.

Read the arithmetic in that sentence.

One serving a day makes a bottle last thirty days; two servings a day makes it last fifteen.

That single choice changes almost everything else on this page.

It doubles the caffeine, doubles every milligram figure, and halves the number of days a bottle covers.

The printed cautions add four more lines, and they are the plainest part of the label.

Don't use it if you're sensitive to any ingredient; don't use it while pregnant, nursing or trying to conceive; ask your clinician first if you take something daily that thins the blood; and there is a caffeine notice, because the capsules carry caffeine.

Four cautions on a supplement label is more than most bottles bother to print.

I'd rather read four than none, and the third one is the reason for a section further down.

Two capsules, away from bedtime. Fifteen or thirty days to the bottle.

How a Panel Like This One Is Actually Read

Before the eight lines, a short detour that will make all eight of them easier.

It is the part nobody explains, and it costs you two minutes.

An American supplement label has two kinds of line on it.

There are the nutrients with an official daily figure beside them, and there are the ones without.

Vitamins and minerals get a percentage, because the government has set a Daily Value for each of them.

Botanicals, amino acids and compounds like the ones on this panel get no percentage, because no such figure exists for them.

So a botanical line can only be read one way: the milligrams on the label against the milligrams in the trial.

That is the whole method of this page, and it is the only fair one.

The second thing worth knowing is the difference between a plant and an extract of a plant.

A gram of dried leaf and a gram of a concentrated extract of that leaf are not the same thing, and the label has to tell you which one it means.

This panel does tell you, on most lines.

It names the plant, the part used, and the extract by its brand name where there is one.

The third thing is the difference between a printed figure and a blend.

A blend prints one number for several ingredients together, which means you cannot know how much of any of them you are getting.

This bottle has no blend on it.

Every one of its eight lines carries its own figure, and that is the single best thing about the label.

Keep those three ideas in your hand for the rest of this page.

A milligram figure, a named extract, and no blend hiding the arithmetic.

Four Point Six From Seventy-Nine Buyers, and the Four Who Wrote Something Worth Quoting

Before the panel, the part most people read first.

The listing shows 4.6 from 79 written reviews: sixty-three at five stars, seven at four, three at three, two at two and four at one.

Every one of them carries the store's own purchase mark, which means the store matched the review to an order.

That is a better-documented set of reviews than most listings on this desk manage, and it still isn't evidence about your memory.

It's evidence that seventy-nine people bought a bottle and most of them were content.

Look at the shape of the distribution for a moment, because the shape says more than the average.

Sixty-three fives and four ones, with almost nothing in between, is the usual pattern for a product people either get on with or send back.

A score built that way tells you very little about the middle of the road.

It tells you there is a group who liked it enough to write, and a small group who did not.

I read the first page of them and pulled the four that said something specific enough to be worth your time.

Two I left out: one was about a parcel that never arrived, which is a shipping story, and one was four words long.

Here they are as written, with the store's purchase mark, and with illustrated portraits rather than photographs of the writers.

John M. Confirmed Buyer

I have been trying various supplements to enhance my cognitive function and memory. This one seems to be helping. I take it in the morning and it helps me get started and thinking clearly throughout the day.

Iman Z. Confirmed Buyer

I have been using them for almost 4 weeks, I believe they help me focus on my tasks.

Jared F. Confirmed Buyer

Great way to get focused on those busy days. I was having issues with feeling overwhelmed and trying to stay on track with daily tasks because of the emergencies that would come flying out of nowhere.

Jorge W. Confirmed Buyer

Love it and it really helps and works.

Read those the way I do.

John says "seems to be helping," and he says he takes it in the morning.

Iman says "almost 4 weeks" and "I believe."

Jared is describing a busy week at work, not a memory.

Nobody in those four sentences describes remembering something they would otherwise have forgotten.

All four describe getting going, staying with a task, and the first part of a day.

Every one of those sentences is about the first hours of a morning, and there are 50 milligrams of caffeine in the serving.

I'd hold that thought until the caffeine section.

Seventy-nine buyers, 4.6, and four careful sentences. A good score, honestly documented, about how people felt, not about what the panel did.

Three Hundred Milligrams of KSM-66, Which Is Exactly the Amount of Its Trial

The first of the three lines that land where their trial landed.

Ashwagandha is a root from an Indian shrub, used in Ayurvedic practice for a very long time.

The version here is KSM-66, a branded root extract, at 300 milligrams per serving.

There is a trial at exactly that figure.

A hundred and twenty-five adults aged 20 to 55 took 300 milligrams a day of a sustained-release root extract for ninety days, and the extract group scored better on memory and focus measures than the placebo group. ^[1]

Ninety days is a long run for a botanical trial, and 125 people is a respectable number for one.

Both of those things are worth saying before the limits, because most of the trials on this desk are shorter and smaller.

Now the limits, and there are three.

The participants were 20 to 55, which is younger than the man this page is written for, and the trial recruited people whose days were demanding rather than a general sample. ^[1]

The third is a form question.

The trial used a sustained-release preparation, and the panel here prints a root extract without saying whether the release is matched. ^[1]

A second, smaller study used a different form again.

Fifty-nine young adults, average age 23, took 225 milligrams a day of a liposomal root-and-leaf extract for thirty days and did better on some word-recall and reaction measures than placebo. ^[2]

That one is in people less than half the age of this page's reader, in a different preparation, so I count it as background rather than support.

The 300-milligram line is the one that matters, and this bottle prints exactly 300.

One more thing about this line, and it is the reason it comes before the ginkgo on my list.

It is a root that people in India have eaten and brewed for centuries, which is not evidence, but is a long record of a kind a new compound does not have.

The listing's own claim for this ingredient uses a word about the way a demanding week feels.

I won't repeat that word here, and I wouldn't lean on it anyway, because the memory and focus measures are the ones with a number attached.

Ashwagandha at 300 milligrams, the dose of a ninety-day trial in 125 adults aged 20 to 55. The clearest match on this panel.

Twelve Milligrams of Lutein and Zeaxanthin, the Figure Two One-Year Trials Used

The second line that lands on its number, and the one most people skip because it sounds like an eye supplement.

Lutein and zeaxanthin are the yellow pigments that sit in the centre of the retina, and they come from marigold, kale, spinach and egg yolk.

This panel prints 10 milligrams of lutein and 2 of zeaxanthin, which is 12 together.

Twelve is the number two one-year trials used.

In the first, sixty-two adults living at home, average age 74, took 12 milligrams a day for a year; the pigment levels in their retinas rose and they did better than placebo on measures of complex attention and mental flexibility. ^[3]

Fifty-one of those sixty-two were analysed at the end, which is the figure to keep. ^[3]

Executive function, the third measure they took, showed only a trend. ^[3]

The second gave 12 milligrams a day for a year to forty-four adults averaging 72, and the supplement group held their ground on a word-learning task while the placebo group slipped. ^[4]

The authors' own phrase was that it "appeared to buffer" the change, which is the careful way to say it, and it's how I'll say it too. ^[4]

That study also ran brain scans during the learning task and saw the pattern of activity change. ^[4]

A scan result is interesting, and it is not the same as remembering more names at a party; the authors did not claim it was.

A third trial is worth knowing because it shows the company this pigment usually keeps.

Sixty adults over 65 took a daily combination of fish oil, 22 milligrams of carotenoids and vitamin E for two years and made fewer errors on working-memory tasks than placebo, with small effects that grew as the tasks got harder. ^[5]

Note what that third trial cannot tell you.

Three nutrients went in together, so nobody can say which of the three, if any, did the work. ^[5]

Two one-year trials, fifty-one and forty-four people analysed, both at 12 milligrams, and this bottle prints 12 milligrams.

That is the second honest match on the panel, and the studies behind it are small.

One thing the listing says that the panel doesn't quite support.

The description says the branded pigment blend supplies meso-zeaxanthin as well, and the printed panel lists only lutein and zeaxanthin, with no figure for a third pigment.

The other claim on this line is about screens and blue light.

I have no trial in front of me that measured that for this dose, so I am not going to pass it along as though I did.

Lutein and zeaxanthin at 12 milligrams, the figure of two one-year trials in adults averaging 72 and 74. Small trials, right number.

Fifty Milligrams of Caffeine in a Capsule, Which Is Half a Cup of Coffee

The third match, and the one the listing does not put in its opening paragraph.

The panel prints 50 milligrams of caffeine per serving, and the printed cautions carry a caffeine notice.

It arrives as part of a co-crystal, a form in which caffeine and trans-pterostilbene are bonded together, which the maker says releases more slowly than caffeine on its own.

Fifty milligrams is a real dose and a modest one.

The Food and Drug Administration puts a 12-ounce cup of brewed coffee anywhere from 113 to 247 milligrams, and cites 400 milligrams a day as an amount not generally tied to negative effects in most healthy adults. ^[6]

Europe's food safety agency reached the same 400-milligram daily figure, and 200 for a single serving. ^[7]

It also says caffeine in moderate amounts makes people more alert and less sleepy, which is the one effect on this panel nobody argues about. ^[7]

It is about half a mug. Count it with your coffee, not instead of it.

Now run the arithmetic the label allows.

Two servings a day, which the direction permits, is 100 milligrams of caffeine on top of whatever is in your cup.

Add two mugs of ordinary brewed coffee to that and you are somewhere between 330 and 590 milligrams for the day. ^[6]

The upper end of that range is above the figure the agency cites, and it got there without anybody doing anything unusual.

The same agency lists what too much caffeine brings: a faster heartbeat, the jitters, an upset stomach, a headache and a poor night. ^[6]

That is why the label says away from bedtime, and why I'd take this bottle in the morning or not at all.

There is one more thing about caffeine in a capsule rather than in a cup.

A mug announces itself: you taste it, you feel the heat, and you know roughly how strong it was.

A capsule announces nothing.

The only way to count caffeine you swallow is to read the panel and write the number down, which is exactly what most people never do.

It also explains something about the reviews.

Four people wrote about getting started in the morning and staying on track through a busy day, and 50 milligrams of caffeine is the line on this panel most likely to be doing that work.

I want to be fair to the bottle here, because that is not a criticism.

An honest, printed, modest caffeine dose with a notice beside it is better behaviour than a blend that hides the same thing behind a plant name.

Fifty milligrams of caffeine, half a mug, with a real short-lived effect on alertness and a notice on the label. Honest, and not the line the listing leads with.

A Hundred Milligrams of Bacopa, Against the Three Hundred of Its Trials

The first line that falls short of its number, and it falls short by two thirds.

Bacopa monnieri is a creeping marsh herb from India, standardised here to 20 per cent bacosides, at 100 milligrams per serving.

Its trials are among the better ones in this whole field, and they all used 300.

In the first, fifty-four adults aged 65 and over took 300 milligrams a day for twelve weeks; forty-eight finished, and they recalled a delayed word list better and did better at ignoring irrelevant information than the placebo group. ^[8]

Nothing moved on divided attention or immediate working memory, and the commonest complaint was an upset stomach. ^[8]

That is a trial in exactly the age group this page is written for, which is unusual and worth noticing. ^[8]

Most of the studies I read in a week were run in undergraduates.

In the second, ninety-eight adults over 55 took 300 milligrams a day for twelve weeks; eighty-one finished, and verbal learning and delayed recall came out ahead of placebo. ^[9]

Other memory and attention measures didn't differ, and the extract group reported more cramps, nausea and trips to the bathroom. ^[9]

Seventeen of the ninety-eight did not finish, and the stomach complaints are the likely reason. ^[9]

That is a real cost of the 300-milligram dose, and it is the honest counterweight to wanting more of it in the capsule.

A 2014 pooled analysis of nine placebo-controlled trials, 518 people in total, all at twelve weeks or longer, found faster performance on two measures of attention speed and asked for a large well-designed trial to settle it. ^[10]

The differences there were measured in milliseconds. ^[10]

Milliseconds on a computer task is what the literature has.

Nobody has shown that it turns into a name arriving when you want it at a dinner table, and the authors of that analysis were careful not to say so. ^[10]

So the ingredient has the best paper trail on this panel, and this bottle carries a third of the amount that produced it.

At two servings a day, the label's upper direction, you reach 200 milligrams, which is still short.

There is a reading of that choice that is sympathetic to the maker.

Less bacopa is less stomach trouble, and a formula with eight lines cannot give every line its full trial dose without becoming a very large capsule.

There is also a reading that is not.

A third of the dose of the ingredient with the best evidence is a third of the dose, however the capsule is sized.

Bacopa at 100 milligrams against the 300 a day of every trial worth citing, in adults 55 and over. Right plant, third of the dose.

What "Standardized to 20 Per Cent Bacosides" Actually Buys You

That phrase sits on the bacopa line, and it deserves a section, because it is the most useful thing on a botanical label and almost nobody reads it.

A plant is not a constant.

The same herb grown in two fields, picked in two months, dried two ways, will carry different amounts of whatever people take it for.

Standardisation is the maker's answer to that.

It means the extract has been adjusted so that a named group of compounds makes up a stated share of the weight, batch after batch.

On this line the named compounds are bacosides, and the stated share is 20 per cent.

So 100 milligrams of this extract carries 20 milligrams of bacosides, and the next bottle should carry the same.

That is worth real money, and it is the difference between a repeatable product and a handful of dried leaf.

It is also the thing that makes the comparison with the trials possible at all.

Now the limit of it, which matters just as much.

A standardised share tells you the extract is consistent; it does not tell you the trials used the same standardisation.

The twelve-week trials in older adults used standardised extracts of their own, one of them a patented preparation. ^{[8] [9]}

Whether 100 milligrams of this one behaves like 100 milligrams of those is not something a label can answer.

So the fair sentence is a careful one.

This is a consistent extract at a third of the amount that the trials used, and "consistent" is not the same as "equivalent".

Apply the same reading to the ashwagandha line.

KSM-66 is a branded, standardised root extract and the ninety-day trial used a branded sustained-release root extract; those are close, and close is the strongest word available. ^[1]

Two Hundred Forty Milligrams of a Complex, and No Figure for How Much Leaf Is Inside It

The line that has a number without telling you what the number counts.

The panel prints 240 milligrams of Ginkgo Phytosome, a branded complex of ginkgo leaf extract bound to sunflower phospholipids.

The maker's case is that binding it to a fat carrier gets more of it absorbed, which is a reasonable thing to build.

Here is the problem with reading it.

Two hundred and forty milligrams of a complex is not 240 milligrams of ginkgo extract, and the label prints no figure for how much leaf extract the complex holds.

Without that figure I can't set this line beside its trials.

I can tell you the arithmetic works against the reader, because the carrier is part of the weight and the carrier is not ginkgo.

The trials are worth knowing either way, and they are the largest in this whole field.

The biggest gave 3,069 adults aged 72 to 96 either 120 milligrams of a standardised ginkgo extract twice a day or a placebo, followed them for a median of six years, and found no difference in memory, attention, language or executive function. ^[11]

Read the size of that one again.

Three thousand and sixty-nine people, six years, 240 milligrams a day, and nothing. ^[11]

A smaller and more favourable trial gave ninety-three adults aged 55 to 79 120 milligrams a day for twelve weeks and found a moderate advantage on one associational-learning measure. ^[12]

Nothing statistically meaningful turned up on any other cognitive, attention or executive measure it took. ^[12]

One positive result among many measures is a thin thing to build on, and the authors said so. ^[12]

The same paper found nothing at all in the 104 young adults it also tested. ^[12]

The federal government's centre for complementary health puts it plainly: there is no conclusive evidence that ginkgo helps any health condition. ^[13]

The same page says ginkgo may raise the risk of bleeding in people who take something daily to thin the blood. ^[13]

That is the line I would name at the pharmacy counter before the first serving, and the label names it too.

The printed cautions ask you to consult your own clinician if you take something daily that thins the blood, because of the ginkgo. ^[13]

I want to be careful about the shape of that sentence.

It is not a rare event in a footnote; it is the maker's own note of caution, printed on its own label, and it is the most important line on it.

There is also a practical version of the same advice.

If anything is scheduled with a date on it at a hospital or a dentist, the bottle is one of the things to mention when they ask what you take.

A 240-milligram complex with no figure for the leaf extract inside it, from a plant whose largest trial in 3,069 older adults came out even. The one line to take to the pharmacist.

What a Phytosome and a Co-Crystal Are, in Plain English

Two words on this panel describe how an ingredient is packaged rather than what it is.

Both are worth a paragraph, because both are doing persuasive work.

A phytosome is a plant extract bound to a fat, usually a phospholipid from sunflower or soy.

The idea is old and sensible: many plant compounds dissolve badly in water and better in fat, and the gut absorbs fat well.

So the maker attaches the extract to a carrier and hopes more of it crosses.

When it works, a smaller amount of extract behaves like a larger one.

That is the case for the ginkgo line, and it is a real argument rather than a marketing one.

What it cannot do is tell you how much extract is there, and that is the figure this label does not print.

A co-crystal is a different idea.

Two compounds are crystallised together into a single structure, so that they dissolve as one thing rather than two.

On this panel, caffeine is crystallised with trans-pterostilbene, and the maker's claim is that the pair releases more slowly than plain caffeine.

That is a claim about timing, not about amount: the amount is still 50 milligrams. ^[6]

Slower release, if it happens, is a mild point in the bottle's favour for a morning capsule.

It would mean less of a spike and a longer, flatter effect, which is what most people over sixty say they want from caffeine anyway.

I have no trial of that pairing in front of me at this dose.

So I record it as the maker's description, not as a finding, which is where it belongs.

Two Hundred Fifty Milligrams of Carnitine and Sixty-Six of Pterostilbene

The last two lines on the panel, and the ones where I have least to give you.

Acetyl-L-carnitine, at 250 milligrams, is a form of a compound your body already makes and uses to move fat into the cell's furnaces.

The maker's own description is that it matters for cellular energy production and nerve function, which is a description of biology rather than a result.

I have no trial on this desk that tested 250 milligrams of it in healthy adults over 55 and measured their memory.

So I won't dress it up. It is a real compound at a small dose, with nothing behind it I can hand you.

It is worth saying what "nothing behind it" does and does not mean.

It does not mean the compound is inert or that nobody has studied it; it means I could not put a trial at this dose in this population in front of you, and I am not going to imply one.

Trans-pterostilbene, at 66 milligrams, is the other half of the caffeine co-crystal.

It is a compound found in blueberries and related to resveratrol.

Its relative has been tested.

Forty-six overweight adults aged 50 to 75 took 200 milligrams of resveratrol a day for twenty-six weeks and retained words over thirty minutes better than the placebo group. ^[14]

That study is not this ingredient and not this dose.

Pterostilbene is on this panel as the other half of the caffeine, and the fair reading is that the caffeine is what it brought with it.

Two lines out of eight in that position is not a scandal.

It is, however, two lines out of eight, and a panel that prints a figure for everything still does not oblige the figure to mean anything.

Carnitine and pterostilbene: real compounds, printed amounts, and no trial at these doses I can put in front of you. Two of the eight.

What the Kitchen Supplies Before Any of This Does

A short section that costs the maker nothing and costs you nothing either.

Two of the eight lines on this panel come out of ordinary food, and it is worth knowing where.

Lutein and zeaxanthin are in kale, spinach, chard and egg yolk, and the deeper the green or the yellow, the more of them there tends to be.

Nobody is claiming a plate of greens is 12 milligrams.

It is usually less, and it arrives with everything else that comes on the plate, which a capsule does not.

Choline is the other one worth a paragraph, and it sits one step away from this panel rather than on it.

In the Framingham families, people whose food carried more choline, which is eggs, milk, meat and beans, scored better on memory tests than those whose food carried less. ^[15]

That is an association across a town, not a result for a bottle, and the people in it were eating rather than supplementing. ^[15]

It is still the cheapest paragraph on this page.

Two eggs, a handful of dark greens and a tin of sardines cost less than any capsule and carry things no panel prints.

That is not an argument against the bottle. It is an argument about where to start.

Six or Eight: the Three Things a Careful Reader Will Spot

Three small discrepancies that I'd rather tell you than leave.

The first is a count.

The description says six studied botanicals and nutrients; the panel prints eight lines, because the caffeine and the pterostilbene arrive in one co-crystal and the lutein and zeaxanthin in one extract.

Neither number is a lie, and I'd still print eight, because eight is what you swallow.

The second is the meso-zeaxanthin in the description with no line on the panel, which I covered in the pigment section.

The third is the audience.

The same listing describes this bottle for people in their twenties and thirties, for gamers, for students, and for aging baby boomers who want to maintain mental agility.^*

A formula addressed to a twenty-five-year-old at a screen and to a sixty-eight-year-old at a desk is doing one of two things.

Either the maker believes the same panel serves both, which is a position, or the copy is written to be bought by everyone, which is a habit.

It also matters for reading the trials.

The ashwagandha trial was run in people aged 20 to 55 and the pigment trials in people averaging 72 and 74, and a panel that addresses both groups at once cannot lean on both sets of numbers at once.

There is one word in that listing I won't reprint, the category word the maker uses for this kind of formula.

It promises a whole category of effect rather than describing a bottle, and no panel of eight lines can carry a promise that size.

One Serving or Two: the Arithmetic That Changes Everything Else

The label allows two servings a day and most people will never think about which one they are doing.

It is the most consequential decision on the bottle, so here are both columns of it.

At one serving a day you take two capsules, and you get the figures on this page as printed.

Three hundred milligrams of ashwagandha, 12 of the two pigments, 100 of bacopa, 50 of caffeine, and a bottle that lasts thirty days.

At two servings a day, every one of those figures doubles.

Six hundred of ashwagandha, 24 of the pigments, 200 of bacopa, 100 of caffeine, and a bottle that lasts fifteen days.

Now read what doubling does to each line, because it is not the same answer for each.

The bacopa gets closer to its trial dose and is still short. The caffeine doubles and moves you nearer the daily ceiling.

The ashwagandha goes to twice the amount of its ninety-day trial, which no trial on this page tested. ^[1]

The pigments go to twice the figure of both one-year trials, which no trial on this page tested either. ^{[3] [4]}

So "more" is not an upgrade at no cost on this panel: it takes two lines past the numbers that were studied and one line nearer a ceiling.

If it were my bottle, I'd take one serving and I'd take it in the morning.

There is a second arithmetic worth doing, and it is about days rather than milligrams.

Sixty days is the return window, and sixty days at one serving a day is exactly two bottles.

At two servings a day, sixty days is four bottles, and the window closes on the first one long before you have finished the fourth.

That is not a trap; it is just a thing to know before the second order.

Sixty Days, a Quarter Kept, and the Rest of the Small Print

The part of a listing I read before I read anything about the ingredients.

Read what that means with a pencil.

Sixty days is a generous window and a quarter is a real deduction, so a bottle you open and dislike costs you a quarter of it plus whatever the return carriage costs.

I say that without complaint, because both halves are printed in plain sight.

A company that prints the restocking figure is easier to deal with than one that prints a warm sentence and puts the figure three pages down.

Here is how I would use the window.

Sixty days is exactly two bottles at one serving a day, which is enough time to know whether you are getting anything and to send the second one back unopened if you're not.

The phone line is the other half of the small print, and it is worth a line of its own.

It answers on weekdays between eight in the morning and eight at night, Eastern time, which is a real range rather than a form on a page.

The rest of the terms are the ordinary ones.

Shipping costs and delivery times are set at the checkout for your address, and the price is the price on the listing on the day you look.

This page prints no price, and that is deliberate.

A price printed here today would be wrong for somebody reading in three months, and the listing is one click away.

Two bottles, sixty days, and a quarter kept if you send one back.

What the Eight Trials Have in Common

Read one at a time, each study on this page sounds reasonable.

Read all of them together and a pattern shows up that no single section makes obvious.

Start with size.

The largest trial cited here is the ginkgo one, with 3,069 people, and it found nothing. ^[11]

Every trial that found something was small.

A hundred and twenty-five for the ashwagandha, fifty-one analysed for the first pigment trial, forty-four for the second, forty-eight and eighty-one finishing the two bacopa trials. ^{[1] [3] [4] [8] [9]}

That is the single most important sentence on this page: the big trial came out even and the small ones came out ahead.

That pattern is common in this field, and it usually says more about small trials than about big ones.

Next, look at who was in them.

Two were in adults averaging 72 and 74, two in adults 55 and over, one in adults 20 to 55, and one in young adults averaging 23. ^{[3] [4] [8] [9] [1] [2]}

Only four of those populations resemble the reader of this page.

That is better than most panels manage, and it is not all eight lines.

Next, length.

Twelve weeks, twelve weeks, ninety days, one year, one year, two years, six years. ^{[8] [9] [1] [3] [4] [5] [11]}

The two one-year trials and the two-year one are the serious end of that list, and they are the pigment trials. ^{[3] [4] [5]}

Nothing here was tested for the length of time a person actually takes a supplement, which is usually years.

Next, what was measured.

Delayed word recall, a colour-and-word task, complex attention, mental flexibility, verbal learning, working-memory errors, milliseconds of attention speed.

Those are laboratory measures, and they are the best measures available.

None of them is "I remembered the name of the man from the golf club before he told me."

And finally, who paid.

Several of the ingredient trials on this page were run or supported by the people who sell the ingredient, which is normal in this field and worth knowing every single time. ^{[1] [2]}

Put all of that together and you get the honest summary of this whole page.

Small trials, short by the standard of a habit, on laboratory measures, some funded by interested parties, and a big one that found nothing.

Three lines at trial dose is the best thing on this bottle. The trials behind those doses are small, and the largest trial on this page found no difference at all.

Seven Things a Bottle of Sixty Capsules Won't Do

• It won't come with a trial of itself. No study has tested this formula; the trials above tested single ingredients in other people.
• It won't bring the bacopa to its trial dose. A hundred milligrams against 300 a day, and 200 at the label's upper direction.
• It won't tell you how much ginkgo leaf extract is in the complex. The panel prints the complex, not the leaf.
• It won't replace the alertness of your coffee with something stronger. Fifty milligrams is half a mug, and it is in the serving.
• It won't act on any condition a clinician has named. It is a dietary supplement, and its label carries the sentence that says so.
• It won't settle the interaction question for you. That is the pharmacist's two minutes, before the first serving.
• It won't come back at full price if you dislike it. Sixty days to ask, and a quarter kept on approved returns.

If that list annoys you, I'd rather you were annoyed now than on the thirtieth morning with an empty bottle and no idea which line did anything.

If You Do Buy It: One Bottle, the Pharmacist First, Mornings Only, and a Notebook

If you're going to try it, try it as a test with a date on it.

• Buy one bottle of sixty. At one serving a day that is thirty mornings, and thirty mornings is a fair test.
• Take the bottle to the pharmacist before the first serving, with everything else you take. Ask about the ginkgo line by name.
• Keep it to one serving a day unless a clinician tells you otherwise. Two servings doubles the caffeine to 100 milligrams.
• Take it in the morning, away from bedtime. The label asks for that, and the caffeine is the reason.
• Count the 50 milligrams with your coffee. Not instead of it, and not on top of a third mug.
• Write one line a day in a notebook. Not "sharper" or "foggier," but what you were doing at ten and how it went.
• Write the same line on the days you forget the capsules. Those days are your own comparison group, and they cost nothing.
• On the thirtieth morning, read your thirty lines beside the thirty from the month before. That's the only trial of this bottle you'll ever get.

Look at that list the way I do.

Seven of the eight things cost nothing at all.

The notebook is the part people skip and the only part that produces information.

Memory is exactly the faculty you cannot use to judge whether your memory changed, which is why it goes on paper.

This is worth a look if:

• You want a panel where every line has a milligram figure beside it, and you'll read them.
• You like that three of the eight land on the number their trial used, and you accept that two have nothing behind them.
• You'll take the bottle to the pharmacist before the first serving and ask about the ginkgo.
• You'll keep it to the morning and count its caffeine with your coffee.

Give it a miss if:

• You take something daily that thins the blood, until the pharmacist has seen the bottle.
• You're sensitive to caffeine, or you already drink three mugs before noon.
• You want the bacopa at the dose of its trials. This is a third of it.
• You find a capsule nine tenths of an inch long hard to swallow, twice at a time.
• You were hoping a bottle would settle a worry that has been with you for a year. That is a conversation, not a capsule.

Who Sells It, and What That Is Worth

The bottle is sold by the company that makes it, and that's the only place this page links to.

That combination is worth more than it sounds, and it is rarer than it should be.

A great many of the bottles that reach this desk are a brand name, a fulfilment centre and a contact form.

It is not the same as a claim about quality, and I won't turn it into one.

It is a claim about reachability: if something goes wrong there is a person on a phone line, in a state you can find on a map.

The price is the one on the maker's listing on the day you look, and this page prints none.

Look at it before you click, and look at the return terms in the small-print section while you're there.

Buy one bottle.

Thirty mornings is a test, and the pharmacist comes first.

Read the panel you get, then read the month you had.

The bottle those eight lines are printed on

Two capsules almost an inch long, with ashwagandha, ginkgo, bacopa, the two pigments of the eye and as much caffeine as half a cup of coffee, taken away from bedtime.

SEE THE MAKER'S LISTING

Sold by Thorne Research, Inc., Summerville, South Carolina · sixty capsules, two to a serving, thirty servings · away from bedtime · 50 mg of caffeine per serving · sixty days to request a return, 25% restocking fee on approved returns

What Readers Ask Me Before the First Bottle

The Whole Bottle, in Sixty Seconds

• A white bottle of sixty capsules from Thorne of Summerville, South Carolina; two capsules to a serving, thirty servings, away from bedtime.
• Ashwagandha: 300 milligrams, exactly the amount of a ninety-day trial in 125 adults aged 20 to 55. ^[1]
• Lutein and zeaxanthin: 12 milligrams together, the figure of two one-year trials in adults averaging 72 and 74. ^{[3] [4]}
• Caffeine: 50 milligrams, about half a mug, with a real short-lived effect on alertness; count it against the 400-milligram daily figure. ^{[6] [7]}
• Bacopa: 100 milligrams against the 300 a day of every trial worth citing in adults over 55. ^{[8] [9]}
• Ginkgo: a 240-milligram complex with no figure for the leaf extract inside it, from a plant whose 3,069-person trial came out even. ^[11] Take it to the pharmacist. ^[13]
• Carnitine and pterostilbene: real compounds at printed amounts, with no trial at these doses to hand you. ^[14]
• No blend on the panel: every line carries its own figure, which is the best thing about this label.
• One serving or two changes everything: two doubles the caffeine and takes two lines past the numbers that were studied.
• Reviews: 4.6 from 79 buyers, all with the store's purchase mark, four quoted here as written.
• Returns: sixty days to ask, 25 per cent restocking fee on approved returns, weekday phone line, return carriage yours.

What Eight Printed Numbers Are Actually Worth

Here's the whole morning in one breath.

A panel with a figure beside every line, which is rarer than it should be, and three of those figures landing exactly where their trial landed.

One good ingredient at a third of its dose, one complex that won't tell you what's inside it, and two lines with nothing behind them I can hand you.

Fifty milligrams of caffeine in a capsule, which is honest and which is also, I suspect, most of what those four reviewers were feeling at nine in the morning.

And a company that prints its address, its phone number, its restocking fee and its caffeine notice, which counts for something on a desk like mine.

Against all that, the sentence I keep coming back to.

The largest trial on this page ran for six years in three thousand people and found no difference, and every trial that found something was small.

If you're a man who wants a panel he can actually read, who'll show the bottle to the pharmacist and keep it to the mornings, that's worth one bottle and thirty lines in a notebook.

Not because of the 4.6.

Because you'll know exactly what's in the capsule, and exactly how far each of the eight lines sits from its trial.

Buy one bottle, take it to the pharmacist before the first serving, keep it to one serving in the morning, count its caffeine with your coffee, and write one line a day. Thirty mornings in a notebook will tell you more than this page can.

Sixty capsules, thirty servings, sold by the company that makes them

Two capsules almost an inch long, with ashwagandha, ginkgo, bacopa, the two pigments of the eye and as much caffeine as half a cup of coffee, taken away from bedtime.

SEE THE MAKER'S LISTING

Sold by Thorne Research, Inc., Summerville, South Carolina · sixty capsules · two capsules one to two times daily, away from bedtime · read the panel on the bottle you receive

P.S. If you take one thing from this page, take the two numbers that decide it.

Three hundred milligrams of ashwagandha, which is the trial's own figure.

A hundred milligrams of bacopa, which is a third of its own.

P.P.S. If the afternoon haze has been with you for a year, and it's there before the coffee too, talk to your own clinician before you buy anything.

A bottle of capsules was never going to answer a question a clinician answers in fifteen minutes and a blood test.