Nonlinear dynamics and breakup of free-surface flows Jens Eggers Universität Gesamthochschule Essen, Fachbereich Physik, 45117 Essen, Germany Surface-tension-driven flows and their tendency to decay into drops have fascinated scientists for over 200 years. Linear stability theory, developed by Rayleigh, Plateau, and Maxwell, governs the onset of breakup. However, recent experiments and technological applications have focused on nonlinear behavior near the breakup singularity. The Navier-Stokes equation governs the dynamics, with universal scaling laws describing the breakup process. At high viscosities, noise-driven instabilities occur, while at low viscosities, capillary waves drive breakup. The author reviews theoretical and experimental developments, highlighting unsolved problems. The formation of drops is common in daily life, science, and technology. Drops result from free-surface motion, but predicting their sizes and dynamics is challenging. Early experiments by Savart and Rayleigh showed that breakup is an intrinsic property of fluid motion, driven by surface tension. Rayleigh's theory, confirmed by experimental data, predicts the typical drop size. The time scale for breakup is determined by surface tension and inertia. Theoretical research on drop formation has focused on linear stability, with Rayleigh's work being foundational. Recent developments include nonlinear simulations and self-similarity theory, which describe the universal behavior near breakup. The dynamics near breakup are independent of the specific setup, and the motion becomes self-similar. The only relevant parameter is the internal length scale, which characterizes fluid properties. Experiments have shown that the dynamics near breakup are universal, with the pinch point being the same for all cases. The appearance of the motion depends on the scale of observation relative to the internal length scale. At low viscosities, the pinch point forms cones, while at high viscosities, it forms thin threads. The universality of the pinch point is a key feature of drop formation. Simulations have been used to study the dynamics of free-surface flows, with boundary integral methods being particularly useful. These methods involve only information about the surface of the fluid and are effective for inviscid, irrotational flows and highly viscous flows governed by the Stokes equation. The evolution of the surface is determined by the velocity potential, which obeys the Laplace equation. The dynamics of free-surface flows are sensitive to the formation of cusp singularities, even in seemingly innocuous flow situations. Surface tension simplifies simulations but also makes the system more sensitive to noise and prone to numerical instabilities. Boundary integral methods are more accurate but neglect viscous or inertial forces, which become important asymptotically. Inviscid, irrotational flow is governed by the Laplace equation, with the velocity potential obeying the Laplace equation. The evolution of the potential follows from the Bernoulli equation, with the pressure on the boundary givenNonlinear dynamics and breakup of free-surface flows Jens Eggers Universität Gesamthochschule Essen, Fachbereich Physik, 45117 Essen, Germany Surface-tension-driven flows and their tendency to decay into drops have fascinated scientists for over 200 years. Linear stability theory, developed by Rayleigh, Plateau, and Maxwell, governs the onset of breakup. However, recent experiments and technological applications have focused on nonlinear behavior near the breakup singularity. The Navier-Stokes equation governs the dynamics, with universal scaling laws describing the breakup process. At high viscosities, noise-driven instabilities occur, while at low viscosities, capillary waves drive breakup. The author reviews theoretical and experimental developments, highlighting unsolved problems. The formation of drops is common in daily life, science, and technology. Drops result from free-surface motion, but predicting their sizes and dynamics is challenging. Early experiments by Savart and Rayleigh showed that breakup is an intrinsic property of fluid motion, driven by surface tension. Rayleigh's theory, confirmed by experimental data, predicts the typical drop size. The time scale for breakup is determined by surface tension and inertia. Theoretical research on drop formation has focused on linear stability, with Rayleigh's work being foundational. Recent developments include nonlinear simulations and self-similarity theory, which describe the universal behavior near breakup. The dynamics near breakup are independent of the specific setup, and the motion becomes self-similar. The only relevant parameter is the internal length scale, which characterizes fluid properties. Experiments have shown that the dynamics near breakup are universal, with the pinch point being the same for all cases. The appearance of the motion depends on the scale of observation relative to the internal length scale. At low viscosities, the pinch point forms cones, while at high viscosities, it forms thin threads. The universality of the pinch point is a key feature of drop formation. Simulations have been used to study the dynamics of free-surface flows, with boundary integral methods being particularly useful. These methods involve only information about the surface of the fluid and are effective for inviscid, irrotational flows and highly viscous flows governed by the Stokes equation. The evolution of the surface is determined by the velocity potential, which obeys the Laplace equation. The dynamics of free-surface flows are sensitive to the formation of cusp singularities, even in seemingly innocuous flow situations. Surface tension simplifies simulations but also makes the system more sensitive to noise and prone to numerical instabilities. Boundary integral methods are more accurate but neglect viscous or inertial forces, which become important asymptotically. Inviscid, irrotational flow is governed by the Laplace equation, with the velocity potential obeying the Laplace equation. The evolution of the potential follows from the Bernoulli equation, with the pressure on the boundary given