Classification of genus two knots which admit a \((1,1)\)-decomposition (Q2405092)

For a \((1,1)\)-knot, which is of Seifert genus two and neither a torus knot nor a satellite knot, it is proved that it is a vertical banding \(K = K_1 \vee_b K_2\) (see below). For \(K_1\), \(K_2\), five possibilities are found, amongst others, the \(K_i\) can be \(2\)-bridge knots or satellite \((1,1)\)-knots, both of genus one, or both can be \(2\)-bridge knots of crosscap number two. By definition, a \((1,1)\)-knot \(K\) in the \(3\)-sphere has a Heegaard splitting torus surface \(T\) meeting \(K\) in exactly two points such that each of the two solid tori bounded by \(T\) intersects \(K\) in a so-called trivial arc. The latter is an arc that can in \(T\) be completed to a closed curve bounded by a \(2\)-disk in the solid torus. The vertical banding operation connects each of the \(K_i\), placed in one of the solid tori, by a twisted band. A final section describes all satellite tunnel number one knots of genus two. The proofs involve Morse theory and the choice of a Seifert surface of minimal genus two with special properties. It is mentioned that all \(2\)-bridge knots as well as all satellite tunnel number one knots are \((1,1)\)-knots of genus two, and all the latter are tunnel number one knots.