Morimoto's tunnel-number conjecture for connected sums of knots

Let K1,K2S3K_1,K_2\subset S^3 be knots, let K1#K2K_1\#K_2 denote their connected sum, and let t(K)t(K) be the tunnel number of a knot KK. For a knot KK, write E(K)E(K) for its exterior. A primitive meridian in a Heegaard splitting (V1,V2)(V_1,V_2) of E(K1)E(K_1) is a meridian curve μE(K1)\mu\subset\partial E(K_1) isotopic to a curve μ\mu^* on the Heegaard surface V2\partial V_2 such that μ\mu^* intersects an essential disk DV2D\subset V_2 in a single point.

Morimoto's conjecture. The knots K1S3K_1\subset S^3, K2S3K_2\subset S^3, and K1#K2K_1\#K_2 satisfy

t(K1#K2)t(K1)+t(K2)t(K_1\#K_2)\leq t(K_1)+t(K_2)

if and only if either E(K1)E(K_1) or E(K2)E(K_2), say E(K1)E(K_1), has a minimal-genus Heegaard splitting (V1,V2)(V_1,V_2) with E(K1)V1\partial E(K_1)\subset V_1 and a primitive meridian.

The conjecture gives a criterion for when the usual upper bound for the tunnel number of a connected sum can be improved by one. The source notes that it was known for the special class of smallish knots, namely knots that do not contain essential meridional surfaces; its general status is not specified here.

Sources & referencesView supporting material

Primary source

Yoav Moriah, “On the intersection of unknotting tunnels and the decomposing annulus in connected sums”, arXiv:math/0211407 (2002).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.