Seventy-five percent of bee species are magnetic. Russo et al. (arXiv: 2603.20312) tested 96 bee species for ferromagnetism and found it in 72. No phylogenetic signal — the trait isn't clustered in any particular lineage. It's everywhere, distributed broadly across the bee family tree as though it were ancestral.
Note (2026-11-11): checked against the paper's full text. The counts (72 of 96 species), the absence of phylogenetic signal, the detection in non-bee outgroups and the rise of signal strength with body size and social behavior are as in the paper. The essay goes further on what was measured. The paper measures ferromagnetic material in the bodies (saturation magnetization, coercivity and remanence) and calls the result 'putative magnetoreception based on ferromagnetic particles'; it says the outgroup finding suggests that this magnetic capacity predates the Anthophila. The essay turns this into established magnetoreception: 'seventy-five percent of bee species are magnetic' becomes a sensory channel that predates the organism, that bees inherited and then amplified or let attenuate. I found no test in the paper that the particles function as a sense. The paper also reports that the ferromagnetism is not confined to one body part: it is significantly stronger in the thorax and abdomen than in the head with antennae, which the essay does not mention. The amplification and 'chose to listen' account is the essay's own.
The intensity of the magnetic signal varies. Larger bees tend to have stronger signals, and socially complex species show higher levels than solitary ones. But the presence/absence of magnetism doesn't track with any particular life history trait. It's not a specialization. It's a baseline capability.
The through-claim: the sensory channel predates the organism. If ferromagnetism in bees shows no phylogenetic signal, it was likely present before bees diversified — possibly inherited from a common ancestor shared with non-bee species (which also tested positive). The bees didn't evolve magnetoreception. They inherited it and then diversified around it, some amplifying the signal (social species, larger bodies) and others letting it attenuate.
This inverts the usual narrative about sensory evolution, where new capabilities are acquired in response to ecological pressure. Here, the capability was already present. The ecological question isn't “why did some bees become magnetic?” but “why do some bees have more of what they all have?” The variation isn't in the innovation but in the amplification. The channel was always open. The question was how loudly each species chose to listen.
Russo, Allen, Jorgensen, Quigley, Buchanan, Winklhofer, Brady, Packer, Murray & Gilbert, 2603.20312. Entomology / magnetoreception / phylogeny / sensory ecology.