On bodies floating in equilibrium in every orientation (Q6157844)

This is a contribution to Ulam's problem 19 from the Scottish Book, which asks: If a convex body \(K \subset {\mathbb R}^3\) made of material of uniform density \(D \in (0, 1)\) floats in equilibrium in any orientation in water, must \(K\) be spherical? The mathematical interpretation of the physical language is as follows: the density of water is taken to be \(1\), the convex body \(K \subset {\mathbb R}^3\) of uniform density \(D \in (0, 1),\) considered to be submerged into water, is said to float in equilibrium in the direction \(\xi\) orthogonal to the water surface if the line \({\mathcal l}(\xi)\) connecting the center of mass of \(K\) and the center of mass of the submerged part is parallel to \(\xi\); it is said to float in equilibrium in every orientation if \({\mathcal l}(\xi)\) is parallel to \(\xi\) for every \(\xi\). \textit{R. Schneider} [Enseign. Math., II. Sér. 16 (1970), 297--305 (1971; Zbl 0209.26502)]. and \textit{K. J. Falconer} [Am. Math. Mon. 90, 690--693 (1983; Zbl 0529.52001)] showed that the answer to Ulam's problem is positive, provided that \(K\) is centrally symmetric and \(D = \frac{1}{2}\). On the other hand, the author has shown in [Ann. Math. (2) 195, No. 3, 1111--1150 (2022; Zbl 1529.52008)] that there are non-centrally-symmetric convex bodies of density \(D = \frac{1}{2}\) that float in equilibrium in every orientation. In the current paper, the author proves and recalls several results which were used in the case \(D = \frac{1}{2}\) in [loc. cit.] and which, the author believes, ``would help to attack the problem for other densities.'' The main result is formulated in terms of the volume \(\delta = D\cdot \mathrm{vol}_d (K)\) of the submerged part of \(K\). It reads: Let \(d\geq 3\), let \(K \in {\mathbb R}^d\) be a convex body and let \(\delta \in (0, \mathrm{vol}_d (K ))\). If \(K\) floats in equilibrium at the level \(\delta\) in every orientation, then, for all hyperplanes \(H\) that cut off the parts of volume \(\delta\) from \(K\), the cutting sections \(K \cap H\) have equal moments of inertia with respect to all \((d -2)\)-dimensional planes \(\Pi\subset H\) passing through the center of mass of \(K \cap H\) and these moments are independent of \(H\) and \(\Pi\). Conversely, let \(K\) have a \(C^1\)-smooth boundary and let the center of mass of \(K\) coincide with the center of mass of the surface of centers, i.e., the locus of the centers of mass of all parts of volume \(\delta\) that are cut off by the cutting hyperplanes \(H\). If all cutting sections \(K \cap H\) have equal moments of inertia with respect to all \((d-2)\)-dimensional planes \(\Pi\subset H\) passing through the center of mass of \(K \cap H\) and these moments are independent of \(H\) and \(\Pi\), then \(K\) floats in equilibrium at the level \(\delta\) in every orientation. This result is also used to provide an alternate proof of the Schneider-Falconer result referred to above.