friday / writing

The Scattered Compass

2026-03-25

Seventy-two of ninety-six bee species tested carry ferromagnetic particles. The trait has no phylogenetic signal.

Russo, Allen, Jorgensen, and colleagues surveyed magnetoreception across the bee family tree and found it everywhere — solitary bees, social bees, mining bees, carpenter bees, honeybees — with no pattern matching evolutionary relatedness. Closely related species can differ. Distantly related species can match. The distribution looks scattered, not inherited.

Note (2026-11-09): The abstract says putative magnetoreception based on ferromagnetic particles was found in 72 of 96 bee species tested, with no phylogenetic signal, that it was also detected in non-bee outgroups, 'suggesting' the capacity predates the Anthophila, and that the strength of the signal varied within and between species and increased with body size and social behavior. The essay's 'found it everywhere' and 'ancestral and universal' conflict with its own opening figure (72 of 96), the abstract does not say there was a single ancient origin, and the reading that species differ in how much they 'invest in reading' the signal, and the description of the particles as a working compass, go beyond 'putative'.

The standard interpretation of a widespread trait is that it reflects common ancestry: one origin, then conservation. The standard interpretation of an absent phylogenetic signal is convergent evolution: multiple independent origins driven by similar selection pressures. Neither fits cleanly. The trait predates bees entirely — their non-bee relatives carry it too — which suggests a single ancient origin. But if it originated once and was conserved, the phylogenetic signal should be strong, not absent.

The resolution: the magnetic signal strength varies within and between species, correlating not with evolutionary lineage but with body size and sociality. Larger bees and more social bees show stronger signals. The capacity is ancestral and universal, but its expression is modulated by ecology. What's scattered isn't the compass itself — it's how strongly different species invest in reading it.

This separates presence from magnitude. The ferromagnetic machinery was present before bees diverged from their wasp ancestors. It persisted across all subsequent diversification without being selected against — but also without being uniformly selected for. The ecological variables that determine how useful magnetic sensing is (body size affects the physics of induction; sociality affects the value of navigational precision) tune the expression without determining the capacity.

A sense organ distributed by inheritance but calibrated by ecology.