Findings and scope
This is a correspondence item, which the journal states is not peer-reviewed (PDF p.1). William Manton argues that the lead isotope figures in Rankin et al. 2005 plot the chocolate data twice and misplot one cocoa powder sample. In his reading, the isotope data do not support the claim that chocolate lead overlaps with lead aerosols, nor that it was added during manufacture (PDF p.1). He also argues that the Nigerian soils were minimally contaminated and that the high lead in bean shells taken directly from the pod suggests contamination after collection (PDF p.2). In their reply, Rankin and Flegal say they are unsure where the extra points in Figures 2 and 4 came from. They state that their conclusions rest on the Table 3 data and maintain that most of the lead in cocoa and chocolate products accumulates during shipping or manufacturing. They report that lead in the products was “up to over 300-fold higher” than in the cocoa bean nibs (PDF p.2). Neither item reports new measurements.
Key numbers
Secondhand values only, recorded as context (PDF p.2).
| Value relayed | As printed | Primary source named |
|---|---|---|
| Lead, average of Nigerian soil samples | 14.2 ppm | Rankin et al. 2005 |
| Lead, US leached ultisols (lowest) | 8 ppm | Helmke 2000, Table 1.4 |
| Lead, US clay-rich vertisols (highest) | 17 ppm | Helmke 2000, Table 1.4 |
Methods and Evidence Fitness
There are no methods; the item is argument about a published study’s figures and inferences. Evidence Fitness is Context only. The numbers are relayed from Rankin et al. 2005 and from a soil chemistry handbook, and the primary sources should carry any measurement. The PDF does not print a DOI for Rankin et al. 2005.
Assessment limits
Both parties declare interests that bear on the dispute. The letter writer was a paid expert witness for the candy companies named in a lawsuit. The original study was part-funded by the American Environmental Safety Institute, the plaintiff in that lawsuit, though the reply states the funding ended before the manuscript was written (PDF p.2). The isotope-ratio arguments cannot be checked from this item. The first page of the PDF also carries an unrelated exchange on dimethylamine borane poisoning, which was not ingested. Nothing in the item concerns milk chocolate specifically, despite the manifest folder.
Provenance notes
The PDF (2 pages, journal pages A274 and A275) was delivered in batch 02 of order HMI-GAP-INGEST-1 on 22 October 2026, filed under the milk-chocolate folder. The SHA-256 matches the batch manifest. The PDF does not print a DOI, so the DOI could not be checked against the PDF. The letter heading “Sources of Lead in Cocoa and Chocolate” matches the manifest title. The PDF also contains the authors’ reply, which is included in this record.
Related evidence
Update history
22 October 2026: Ingested from the batch-02 PDF. 3 secondhand soil lead values were added to the evidence register as context only.
Update history
No substantive edit history is available in this build. The full commit record is available in git.