Eating Brazil Nuts for Selenium? Can You Rely on Counting Them?

You pick out a single Brazil nut, then look up how much selenium it contains. A nutrition table gives you a number, perhaps even a decimal, and the question seems settled. But before treating that number as a description of the nut in your hand, what would you want to know about it? Where did the figure come from, and what does the table mean by 'one nut'?

The reference number begins with a weight of food. The USDA’s Standard Reference entry for dried, unblanched Brazil-nut kernels gives a mean selenium concentration of 1,917 micrograms per 100 grams. Applying that average to its nominal 5-gram kernel gives 95.9 micrograms. The decimal belongs to a conversion from a reference average, not to a chemical measurement of an individual nut. “One kernel” here represents an assumed edible weight as well as a count.¹

Weight and concentration contribute different information. Concentration describes selenium per mass of nut; the amount in a portion depends on both that concentration and how much kernel is present. Weighing the edible portion makes the mass more specific. Multiplying that weight by the table’s average concentration still leaves an assumption about how closely the actual kernels resemble the reference food. It improves one part of the estimate without measuring the other.^1,2

Sharing a bag does not ensure that kernels share the same selenium concentration. Among thirteen kernels examined from each of two US retail bags labeled Brazil, concentrations varied within the sampled bags. In one bag, the measured range was 10.0 to 78.7 milligrams per kilogram. Those are concentrations, not amounts in whole nuts. Because whole-kernel weights were not reported, that range cannot be translated into the same fold difference in selenium per nut. Nor does this small sample establish the extremes for Brazil nuts across the retail market.²

The growing context also qualifies how confidently a generic average can describe particular nuts. A survey of 71 trees at five selected Brazilian sites found differences associated with the sampled growing sites. It used three nuts per tree for laboratory preparation, rather than a representative sample of retail portions. Site, tree and management differences were not experimentally separated. The site differences show why a generic average may fit some nuts better than others. They do not turn a country or regional label into a selenium measurement.³

For a batch-specific estimate, the tested sample would need to represent the nuts in that batch. A scale supplies weight, but composition data must come from the food itself. The reference table can still provide an estimate, as long as its average is not presented as a measurement of the kernels on the plate.^1,2,3

The number from a nutrition table has a specific role: estimating the selenium in a defined portion. Interpreting it requires keeping the reference average separate from the composition of a particular nut. A correct conversion between units does not remove that distinction. The useful level of precision is the one supported by the food-composition information behind the calculation.