Veins can carry ore minerals and can seal fluid pathways for hydrocarbons and are therefore of economic importance. Veins are a proxy to the conditions that led to their formation and they are structural markers that contain information on later geological activity. The aim of this thesis is to improve our understanding of fracture and seal processes leading to the formation of veins and the effect of these processes on subsequent reactivation or deformation of veins. In particular we want to understand how the strength and elastic behaviour of vein material and vein walls controls a series of developing veins and how this behaviour affects the interaction of later fracture and sealing events. We also investigate how such fracture and sealing processes affect deformation in low-permeability argillaceous rocks and how they create and maintain transient permeability structures during later deformation. To tackle this problem, a dual approach was followed, on the one hand, through a study of the veining at the Nkana copper and cobalt (Cu−Co) deposit in Zambia and, on the other hand, through numerical fracture and sealing modelling experiments using the Discrete Element Method (DEM). The Central African Copperbelt in Zambia and D.R. Congo is the largest and highest-grade sediment-hosted Cu−Co producing metallogenic province in the world. The Nkana deposit is situated in the Chambishi-Nkana structural basin in the eastern part of the Zambian Copperbelt. Sulphide mineralisation in the Copperbelt Orebody Member of the Katanga Supergroup occurs disseminated in the host rock, in nodules and several generations of veins. First a basin-scale structural analysis of the southeastern part of the Chambishi-Nkana basin is carried out to refine existing geodynamic models and to better define the structural framework at the Nkana deposit. Lateral basin-scale lithofacies variations in the Copperbelt Orebody Member are identified in this study, ranging from carbonaceous mudrocks to argillaceous dolomites. Deformation in the Chambishi-Nkana basin is mainly by folding on multiple scales. This folding and tectonic cleavage development is interpreted to have occurred during single NE-SW oriented shortening in the Pan-African Lufilian orogeny. Fold geometries are non-cylindrical as NW dipping periclinal folds often arranged en echelon and interfering laterally. A clear strain gradient, increasing from NW to SE, is observed at the southeast side of the Chambishi-Nkana basin. This gradient is interpreted as the result of, on the one hand, an inclined megascale fold trend with abundant parasitic folding intersected by a horizontal topography and, on the other hand, gradual changes in lithology of the Copperbelt Orebody Member from NW to SE. Abundant veining is observed in the carbonaceous mudrock lithofacies at Nkana. In total nine distinct vein generations can be identified, that are associated with various stages during progressive folding and cleavage development as a result of the Pan-African Lufilian orogeny. The study then focuses on geometrical, (micro)structural and geochemical investigation of two vein generations at Nkana: Barren bedding-parallel fibrous dolomite veins of type I and bedding-parallel elongate−blocky to blocky dolomite-quartz veins with abundant Cu-Co mineralization of type II. Measurements of type I vein spacing and thickness distributions show that these veins are closely spaced and form up to ten volume percent of the host rock. The veins show a continuum between antitaxial and unitaxial growth morphologies. Microstructural investigation clarify that the veins have very smooth vein walls and that dolomite fibres show a degree of growth competition. The veins are interpreted to have formed as horizontal extensional veins with many growth increments, always located on the vein walls. The observed spacing distribution of type I veins indicates a minor degree of clustering on top of an essentially random process. The spacing distribution of the veins could be explained either by the existence of very small differences in the stress−state and tensile strength between vein material, vein wall and host rock or by the effect of stress shadows around open fractures in the Copperbelt Orebody Member prohibiting failure of nearby type I vein walls. The type I veins were most likely formed during burial diagenesis before significant compressive tectonic stresses were applied to the rocks. The horizontal extensional hydrofractures leading to type I veins were most probably formed due to seepage forces by fluid overpressuring in an effective extensional Andersonian stress regime (in which the minor principal effective stress axis is vertical). They are interpreted to be crack-seal veins with incremental openings that were of such a size to cause a limited amount of growth competition in the fibres. Type II veins are intimately related to folding of type I veins and contain abundant Cu−Co ore mineralisation. These veins are bedding-parallel with slickenfibres over elongate-blocky to blocky growth morphologies. The growth morphologies and deformation microstructures in type II veins can be correlated to different stages during progressive folding of the veins. We investigate how exactly the deformation of pre-folding type I veins leads to significant enrichment of the ore in transient permeability structures related to folding of these veins, as exemplified in the type II veins. High viscosity contrasts of type I veins with the mudrocks has led to high amplitude single- and multilayer folding in poly- to disharmonic sequences. The dense, closely spaced populations of type I veins then resulted in multilayer folding and space accommodation problems during fold lock-up, creating transient permeability structures and intense Cu−Co sulphide mineralisation in these zones. The internal strain distribution in folded type I veins is diagnostic of a high contribution of flexural flow. This strain was accommodated mainly through intergranular bookshelf rotation and in minor amounts through intracrystalline bending of initially orthogonal dolomite fibres. Flexural flow folding was thus caused by a planar mechanical anisotropy initially perpendicular to the boundaries of the single-layers, highlighting that the complexity of internal fabric in veins should be taken into account in models of folding. As a second part of this work, a parametric numerical study was carried out using the Discrete Element Method (DEM). From a methodological point of view, a numerical laboratory is developed that allows the deformation of isotropic and anisotropic numerical rock analogues under changing stress-states. The DEM model captures the brittle-elastic behaviour of rock reasonably well. By introducing layering or texture in the numerical model, an anisotropy in mechanical response can be reproduced that is similar to that in real rocks. Several fracture and sealing numerical experiments are conducted. In these experiments, we systematically vary the competence of host rock, vein and vein wall while repeatedly fracturing and sealing the model. A first series of experiments concerns a classic three-layer problem of a competent layer in an incompetent matrix. Secondly, these experiments are repeated on numerical specimens with a layered transversely isotropic matrix. In a third experiment, the effects of a rotation in the remote stress field on fracture processes in a transversely isotropic rock are studied. The model results indicate that the relative strength of host rock, vein material and vein walls determines what the morphology of the developing vein system will be. If veins and vein walls are stronger than host-rock domains, braided vein patterns originate in the numerical experiments. Conversely, crack-seal or ataxial veins are obtained if the vein and vein wall strength is low relative to that of the host rock or when the vein walls are weak relative to other parameters in the model. Most importantly, a transition can be observed in the vein patterns, changing from localised composite veining to delocalised, braided vein patterns depending on the vein strength and critical competence contrasts are obtained. The tensile strength of the different components is hence a strong determining factor of the morphology of the developing vein system. Spaced distributions of crack-seal veins are also observed in the DEM experiments. Such spaced development of crack-seal veins is only observed when differences in tensile strength of the vein and host rock material is very small, leading to creation of a spaced network of crack-seal veins, as for example seen in type I veins at Nkana. nrpages: 236 status: published