Cacao Compounds: What The Research Has Actually Measured
Cacao contains two methylxanthines, caffeine and theobromine, along with flavanols and trace quantities of a family of fatty acid ethanolamides. Those are the compounds the published chemistry keeps returning to, and they are documented in measured amounts. What the research has not produced is an answer to the question people actually type, which is why does cacao make me feel good: the trials are small, short, and measure test scores rather than anything resembling a day.
The gap between those two things is where most chocolate writing lives. A compound gets identified in a laboratory, the finding travels, and by the third retelling it has become a mechanism for an experience nobody measured. Both of the best-known cacao stories went exactly that way, and one of them was publicly corrected in print.
The methylxanthines, and the trial that isolated them
Smit, Gaffan and Rogers published two double-blind, placebo-controlled studies in Psychopharmacology in 2004, with 20 and 22 participants. In the first, they gave 11.6 g of cocoa powder, or a combination of 19 mg of caffeine and 250 mg of theobromine, amounts the authors took from a 50 g dark chocolate bar. Participants did a reaction time task, a visual information processing task and a mood questionnaire, once before treatment and twice afterwards.
Their result was that the cocoa powder and the caffeine-plus-theobromine combination produced the same changes against placebo on an energetic arousal mood scale and on the cognitive tasks, which is how they arrived at the title: methylxanthines are the psycho-pharmacologically active constituents of chocolate. In the second study, milk and dark chocolate doses differed from white chocolate, which did not differ from water.
The limitations are substantial and the authors state one themselves. These are acute single-dose laboratory sessions with roughly twenty people each, measuring questionnaire and reaction-time scores in the hours afterwards. Nothing about repeated use, and nothing about anybody's week. The authors also wrote that other attributes of chocolate are probably more important than methylxanthines for its appeal and for chocolate cravings.
The anandamide story, and the correction
In 1996 a Nature report identified brain cannabinoid-like compounds in chocolate, and the phrase bliss molecule entered circulation and never left. Two years later, in the same journal, the finding was challenged.
Di Marzo, Sepe, De Petrocellis, Berger, Crozier, Fride and Mechoulam published a scientific correspondence in Nature on 17 December 1998, with a reply from Beltramo and Piomelli. Their measurements in cocoa-derived samples found N-acylethanolamines at 0.01 to 5.8 mg/g and oleamide at 0.17 to 6.0 mg/g, but anandamide at less than 0.003 mg/g in cocoa powder, with no 2-AG detected at all. They suggested the tiny anandamide amounts might be processing artefacts, noting that these compounds were much lower in unfermented beans than in cocoa powder.
Their conclusion was arithmetic rather than rhetorical. Estimating that only about 1.6 to 5 per cent of these compounds given orally reach the bloodstream, they wrote that food levels sit several orders of magnitude below what would be needed for central effects. They were careful: their behavioural tests were a gross assessment, and they said subtler work would be needed before ruling out a mild effect entirely. Our page on the anandamide finding in cacao goes through that exchange in more detail.
What has been measured, compound by compound
| Compound | What was measured, and by whom | What the measurement does not establish |
|---|---|---|
| Caffeine and theobromine | 19 mg caffeine and 250 mg theobromine, the amounts attributed to a 50 g dark bar, given to 20 adults (Smit et al., 2004) | Anything beyond acute laboratory scores in about twenty people |
| Anandamide | Less than 0.003 mg/g in cocoa powder (Di Marzo et al., 1998) | A route to the brain: the authors put food levels orders of magnitude below what oral dosing would require |
| N-acylethanolamines and oleamide | 0.01 to 5.8 mg/g and 0.17 to 6.0 mg/g in cocoa samples (Di Marzo et al., 1998) | That they survive digestion; estimated oral bioavailability was 1.6 to 5 per cent |
| Flavanols | 0.095 to 3.264 mg/g across 41 commercial chocolates; epicatechin 32.9 per cent of total flavanols (Alañón et al., 2016) | Anything about a person: the study analysed chocolate, not eaters |
What nobody has measured
Chaski has no measured flavanol figure, no measured methylxanthine figure, and has not been through a trial of any kind. We would rather say that plainly than borrow a number from somebody else's chocolate, because the Alañón analysis showed that chocolates can differ substantially from one another even where their labels do not.
So treat any page that moves from a compound to a promise as having skipped a step it could not take. The compounds are real and the quantities are published. The leap is the invented part. For the methylxanthine comparison people ask about most, see our note on theobromine and caffeine in cacao.
What is actually in a Chaski tube
Chaski Cacao is a 600 g tube of 77 per cent dark chocolate buttons, with four buttons as the daily dose printed on the pack. The cacao is single-origin Peruvian and Ecuadorian, from a farm our founder has visited, ground at low temperature. The sweetener is coconut sugar rather than cane sugar, and the formulation contains no emulsifiers, no seed oils and no palm oil. Every batch is tested for lead, cadmium and pesticides.
Composition, sourcing and testing are the whole of what we will put our name to. The ingredient list and the format are on the Chaski Cacao product page, and the rest of the library sits in our guides.
If you want to go further than this page, read the Di Marzo correspondence first. It is two pages long and it shows a famous finding being argued with, which is rarer in this category than it should be.
Frequently Asked Questions
What compounds does cacao contain?
The published chemistry keeps returning to a short list. Cacao contains the methylxanthines caffeine and theobromine, measured by Smit and colleagues in 2004 at 19 mg and 250 mg respectively for the amounts they attributed to a 50 g dark chocolate bar. It contains flavanols, measured by Alanon and colleagues in 2016 at 0.095 to 3.264 mg/g across 41 commercial chocolates, with epicatechin making up 32.9 per cent of the total. It also contains trace fatty acid ethanolamides, including anandamide at less than 0.003 mg/g in cocoa powder.
Is the bliss molecule story about chocolate accurate?
It was challenged in print. A 1996 Nature report identified brain cannabinoid-like compounds in chocolate. In Nature on 17 December 1998, Di Marzo, Sepe, De Petrocellis, Berger, Crozier, Fride and Mechoulam published measurements finding anandamide at less than 0.003 mg/g in cocoa powder and no 2-AG, suggested the trace amounts might be processing artefacts, and estimated that only about 1.6 to 5 per cent of such compounds given orally reach the bloodstream. They concluded that food levels sit several orders of magnitude below what central effects would require. Beltramo and Piomelli replied in the same issue.
What did the 2004 chocolate and methylxanthine trial actually do?
Smit, Gaffan and Rogers ran two double-blind, placebo-controlled studies published in Psychopharmacology in 2004, with 20 and 22 participants. In the first, participants received 11.6 g of cocoa powder or a combination of 19 mg caffeine and 250 mg theobromine, with a reaction time task, a visual information processing task and a mood questionnaire administered once before and twice after. The authors reported the cocoa powder and the methylxanthine combination producing the same changes against placebo, and concluded the methylxanthines accounted for them. Both studies were acute single-dose laboratory sessions.
Are there measured compound figures for Chaski Cacao?
No. There is no measured flavanol figure, no measured methylxanthine figure and no trial of the product. We would rather state that than borrow a number from another chocolate, particularly since the 2016 analysis of 41 commercial chocolates found no positive correlation between flavanol content and the percentage of cocoa solids on the label. What is documented for the tube is composition, sourcing and batch testing: 600 g of 77 per cent dark buttons, single-origin Peruvian and Ecuadorian cacao, coconut sugar, and every batch tested for lead, cadmium and pesticides.
Sources
- Smit HJ, Gaffan EA, Rogers PJ, Psychopharmacology, 2004
- Di Marzo V, Sepe N, De Petrocellis L, Berger A, Crozier G, Fride E, Mechoulam R; reply by Beltramo M and Piomelli D, Nature, 1998
- Alañón ME, Castle SM, Siswanto PJ, Cifuentes-Gómez T, Spencer JPE, Food Chemistry, 2016