The stable copper isotope composition of 79 samples of primary and secondary copper minerals from hydrothermal veins in the Schwarzwald mining district, South Germany, shows a wide variation in δ 65Cu ranging from −2.92 to 2.41‰. We investigated primary chalcopyrite, various kinds of fahlores and emplectite, as well as supergene native copper, malachite, azurite, cuprite, tenorite, olivenite, pseudomalachite and chrysocolla. Fresh primary Cu(I) ores have at most localities copper isotope ratios ( δ 65Cu values) of 0 ± 0.5‰ despite the fact that the samples come from mineralogically different types of deposits covering an area of about 100 by 50 km and that they formed during three different mineralization events spanning the last 300 Ma. Relics of the primary ores in oxidized samples (i.e., chalcopyrite relics in an iron oxide matrix with an outer malachite coating) display low isotope ratios down to −2.92‰. Secondary Cu(I) minerals such as cuprite have high δ 65Cu values between 0.4 and 1.65‰, whereas secondary Cu(II) minerals such as malachite show a range of values between −1.55 and 2.41‰, but typically have values above +0.5‰. Within single samples, supergene oxidation of fresh chalcopyrite with a δ value of 0‰ causes significant fractionation on the scale of a centimetre between malachite (up to 1.49‰) and relict chalcopyrite (down to −2.92‰). The results show that—with only two notable exceptions—high-temperature hydrothermal processes did not lead to significant and correlatable variations in copper isotope ratios within a large mining district mineralized over a long period of time. Conversely, low-temperature redox processes seriously affect the copper isotope compositions of hydrothermal copper ores. While details of the redox processes are not yet understood, we interpret the range in compositions found in both primary Cu(I) and secondary Cu(II) minerals as a result of two competing controls on the isotope fractionation process: within–fluid control, i.e., the fractionation during the redox process among dissolved species, and fluid–solid control, i.e., fractionation during precipitation involving reactions between dissolved Cu species and minerals. Additionally, Rayleigh fractionation in a closed system may be responsible for some of the spread in isotope compositions. Our study indicates that copper isotope variations may be used to decipher details of natural redox processes and therefore may have some bearing on exploration, evaluation and exploitation of copper deposits. On the other hand, copper isotope analyses of single archeological artefacts or geological or biological objects cannot be easily used as reliable fingerprint for the source of copper, because the variation caused by redox processes within a single deposit is usually much larger than the inter-deposit variation.