The research.

Each page in this section takes one question about honey, another bee product or olive oil and reads the published research on it properly: the trials in full — who was studied, what was measured, what the numbers were — the laboratory work on its own terms, and the systematic reviews that weigh everything together. Where the research found something, the page says so in its first sentence. Where it found nothing, it says that just as plainly, because an empty record is a finding too, and often the more surprising one. Every number is cited to its source.

How to read an evidence level

Not every finding is the same kind of fact. A result in a laboratory dish — in vitro — shows what a substance can do to cells or microbes under controlled conditions; it says nothing yet about what happens in a person. An animal study is a step closer. A human trial is the real test, and the randomised controlled trial, where chance rather than choice decides who receives what, is the strongest single design: randomisation is what separates an effect from a coincidence.

Above any single trial sits the systematic review, which gathers every eligible trial on a question and weighs them together, then grades its own conclusion. The grades are worth knowing: high-certainty evidence means further research is unlikely to change the answer; moderate means it well might; low means the number is closer to a hint than a measurement. The pages here name the design and the grade inline, every time, because the kind of study decides what a result can mean.

What every page does

• Reads the primary source — findings come from the papers themselves, not from coverage of them; quoted words are the authors' own.
• Reports the numbers — participant counts, effect sizes and confidence intervals, so you can judge the size of a result and not just its direction.
• States the limits once, where they matter — a small unblinded study is called small and unblinded in the sentence that reports it, and the reading moves on.
• Keeps the null results — a trial that found nothing is reported with the same care as one that found something; knowing what failed is half of knowing a subject.
• Includes the regulators' verdicts — where Europe's food-safety assessors and the Great Britain register have judged the same evidence, the pages report how and why, because a panel's published assessment is a close reading in its own right.
• Keeps the safety facts in view — bee-product allergy is real and occasionally serious, and honey is not for infants under 12 months; each page carries the detail where it applies.

The pages

One question per page: what has been studied, at what level, and what was actually found — the hive first, then the grove.

The hive — honey and bee products

The grove — olive oil

Why read it this way?

Because almost everything written about honey — and nearly as much written about olive oil — is a retelling of a retelling, and the interesting material — the trial that surprised its own authors, the review grade that shifted a conclusion, the question nobody has actually tested — lives in the primary record. An hour with these pages leaves you knowing what has genuinely been measured, and the habit they teach — measured how, on whom, against what? — transfers to every other subject you read about. The same standard runs through the rest of the house: how each batch is tested and where everything comes from.