Decomposition rates are an important component of carbon sequestration rates in soils, potentially mitigating future climate change. Here we aim to better understand decomposition's relationship with temperature in natural conditions. In snow-tussock grassland dominated by Chionochloa rubra on Mount Tongariro, Tongariro National Park, New Zealand, we measured decomposition of Chionochloa leaf litter along an ≈ 700 m altitudinal gradient, as a space-for-temperature experiment, representing 4.2°C of warming. We examined decomposition rates in a full reciprocal translocation of litter bags between eight plots as both the origin of eight litter types and the eight destinations of plating out of litter bags, over 4 years using six replicates, and modelled their relationships to environmental variates. Litter decomposed progressively over time, but at the same rate along the altitudinal gradient. There was no home-field advantage. In terms of litter quality, decomposition rates were related only to litter lignin, or fibre or litter N. Only decomposition at Year 4, and that only when organised by litter destination, showed a relationship to mean annual temperature jointly with soil C, and this was only weak and implausible. When studied across the full reciprocal transplant, there were no significant interactions between Origin and Destination data with or without Years. Therefore litter from each plot decomposed at the same rate as other plots' litter at all altitudes, allowing for small, often irregular differences in litter quality and micro-environment. Despite the few modelled differences, decomposition rates show no plausible trends in our altitude-for-temperature substitution. We suggest this may be a universal finding, except perhaps under different moisture regimes. Thus, under projected climate warming scenarios, changes in temperature will not directly affect decomposition rates, and cannot influence C sequestration in nature. Read the free Plain Language Summary for this article on the Journal blog.