← All problems

Hilbert’s 16th problem, second part

For each integer d≥1d\ge 1, let H(d)H(d) be the supremum of the numbers of limit cycles of planar polynomial systems x˙=P(x,y)\dot{x}=P(x,y), y˙=Q(x,y)\dot{y}=Q(x,y) satisfying max⁡{deg⁡P,deg⁡Q}≤d\max\{\deg P,\deg Q\}\le d. Determine whether H(d)<∞H(d)<\infty for every dd; equivalently, whether there is a uniform finite upper bound, depending only on dd, for the number of limit cycles of all planar polynomial systems of degree at most dd.

References

Progress summary

Refreshed
Open · progress reported

The problem remains open, but a new paper claims a stronger lower bound on how many repeating motions these systems can have.

The problem asks whether systems of a fixed degree can have a uniform finite upper bound on their number of limit cycles, denoted H(n)H(n). Finiteness for each individual system is known, but the uniform-bound question remains open.

Known results

  • Dulac (1923) proved finiteness for each individual polynomial vector field.
  • Chen, Wang, and Shi (1979–80) constructed quadratic systems with four limit cycles, proving H(2)≥4H(2)\ge 4.
  • Écalle and Ilyashenko (1991–92) independently repaired the gap in Dulac’s finiteness proof.
  • For general degree, the known lower-growth behavior is H(n)H(n) growing as n2log⁡nn^2\log n.

October 8, 2026 claimed lower-bound improvement

Chaoyang Qin and Xiaoming Sun’s preprint claims a logarithmic-factor improvement by constructing many limit cycles across multiple period annuli. This is a claimed partial advance, not an independently verified resolution; the general finiteness and upper-bound questions remain open.

Current status (as of October 2026): Individual-system finiteness is settled, but the uniform upper-bound problem remains open; a logarithmic-factor lower-bound improvement is claimed by a preprint and is unverified.

Sources

Solutions 0

No solutions have been posted yet.