From the covariant bound on the entropy of partial light-sheets, we derive a version of Bekenstein's bound: S/M ≤ πx/h, where S, M , and x are the entropy, total mass, and width of any isolated, weakly gravitating system. Because x can be measured along any spatial direction, the bound becomes unexpectedly tight in thin systems. Our result completes the identification of older entropy bounds as special cases of the covariant bound. Thus, light-sheets exhibit a connection between information and geometry far more general, but in no respect weaker, than that initially revealed by black hole thermodynamics.Entropy bounds have undergone a remarkable transformation from a corollary to a candidate for a first principle [1]. After proposing the generalized second law of thermodynamics (GSL) [2,3]-that the sum of black hole entropy and ordinary matter entropy never decreasesBekenstein argued that its validity necessitates a modelindependent bound [4,5] on the entropy S of weakly gravitating systems:where M is the total gravitating energy, and d is the linear size of the system, defined to be the diameter of the smallest sphere that fits around the system. This inequality is obtained by considering the classical absorption of the system by a large black hole; it does not depend on the dimension of spacetime [6]. Bekenstein's bound is remarkably tight (consider, for example, a massive particle in a box the size of its Compton wavelength). It has appeared in discussions ranging from information technology to quantum gravity. Since M ≪ d/4G for a weakly gravitating system, it also implies the "spherical entropy bound",Here A cs is the area of the circumscribing sphere. Though confined to weak gravity, 't Hooft [7] and Susskind [8] ascribed fundamental significance to Eq. (2), claiming that it reflects a non-extensivity of the number of degrees of freedom in nature. This eventually prompted the conjecture of a more general bound, the covariant entropy bound [9]. Empirically, this bound has been found to hold in large classes of examples, including systems in which gravity is the dominant force. Meanwhile, no violation has been observed, nor have any theoretical counterexamples been constructed from a realistic effective theory of matter and gravitation.Although the covariant bound does not conflict with the phenomenology of our present models, it cannot be derived from known principles. It may be interpreted as an unexplained pattern in nature, betraying a fundamental relation between information and spacetime geometry.Then the bound must eventually be explained by a unified theory of gravity, matter, and quantum mechanics. In the mean time, it should be regarded as providing important hints about such a theory.We are thus motivated to consider the covariant entropy bound primary, and to try to derive other laws of physics from it. As we will shortly discuss, the bound has already been shown to imply the GSL, as well as older, more specialized entropy bounds. However, the oldest (and, for weakly gravitating systems, tightest...