Robert M. May (1972), Nature 238(5364), 413–414 — DOI-10.1038/238413a0
Key takeaways
Filed, not read. Known here by what it is standardly cited for.
- For a random community matrix with n species, connectance C and interaction strength s, the system is almost surely stable if s√(nC) < 1 and unstable if s√(nC) > 1.
- The transition is sharp as n grows, so increasing size or connectance past a threshold destabilizes the system.
- It challenged the then-common view that ecological complexity confers stability, and started a long debate over which structures escape the bound.
Standing in the record
Filed on 2026-10-11 at the owner’s request. It bears on flooding (CLAIM-179): if anchors are components and the relations between them are interactions, a record that keeps adding entries and links may pass a threshold past which it no longer settles. The record’s structure is not random, and the debate after May is about exactly which structures escape the bound, so the analogy needs care. The record holds the ecological side of that debate (Holling, LIT-741; THEORY-132), not May himself until now.
Unavailable since 2026-10-14. No lawful full text was reached, so the threshold s√(nC) < 1 quoted above and in CLAIM-179 is the paper’s standard citation, not checked against its text here. CLAIM-179 already marks it unread.