LAUSR

that, although the hypothesis of ancient symbiotic events leading to transfer of magnetite biomineralization genes (MBGs) from magnetotactic bacteria (MTB) to eukaryotes has been raised for decades (2), Bellinger et al. (3) do not provide evidence supporting that MBGs are, per se, functionally equivalent to magnetosome genes homologs. Obvious hypothesis precursors include quantifying whether eukaryote genomes contain the neces-sary genetic machinery, that is, suites of distant homologs of MTB magnetosome biomineralization genes, and associating those with magnetite production. Accordingly, we (3) exam-ined genomes of 13 phylogenetically diverse eukaryotes, many known for keen navigational sense, and the Asgard archea clade Lokiarchaeota. If MBGs were absent in some or all, then production of biogenic magnetite across diverse life forms could only be explained by convergent evolution. However, our fi ndings point to a different path in support of ancient symbiotic events: Distant homologs of MBGs are universally present in eukaryote genomes (and Lokiarchaeota), including four of fi ve core genes universally shared by MTB Nitrospirae and Proteobacteria (4), composing a subset of the minimal set of genes required for magnetosome biomineralization in prokaryotes (5, 6). Our genetic hypothesis test-ing was then extended to include transcriptomics data from candidate magnetite-based magnetoreceptors contained in salmonid olfactory tissues (7). That distant homologs of MBGs were differentially and more highly expressed in magnetic relative to nonmagnetic olfactory cells is compelling evidence for our bold, yet not unprecedented (2),