A line field on a surface is a smooth map that assigns a tangent line to all but a finite number of points. Such fields model a number of geometric and physical properties, e.g. the principal curvature directions on surfaces or the stress flux in elasticity. They can be seen as a generalization of vector fields. To understand a line field, it is common to study the behavior of its orbits, which can have many different patterns. To this end, we consider a topological approach: we use the critical points and separatrices to decompose the field in regions of similar behavior. We focus on fields that have a Morse–Smale structure. This allows operations like the cancellation of critical points controlled directly in the field decomposition, which is essential for noise removal (topology simplification) on fields coming from simulations or sampling of real-world problems. Based on the decomposition of a Morse–Smale vector field and on cancellation of critical points, Robin Forman introduced a discrete definition for Morse-Smale vector fields. This thesis provides a purely combinatorial definition of line fields, the discrete line fields, entailing Forman s discrete constructions for vector fields through a new representation of these. Discrete line fields admit a (Morse–Smale type of) decomposition that generates a bridge between discrete and smooth line fields, thus guaranteeing the topological consistency of the definition. We also use double branched coverings to suspend discrete line fields to discrete vector fields, so that vector field tools can be used for discrete line fields. Finally we provide, for a discrete line field, a topologically consistent (Morse-like) cancellation of critical elements. This allows a simplification of the discrete line field topology retaining only the most significant features.