Carbon isotope ratios in higher–plant organic matter (δ13Cplant) have been shown in several studies to be closely related to the carbon isotope composition of the ocean–atmosphere carbon reservoir, and, in particular, the isotopic composition of CO2. These studies have primarily been focused on geological intervals in which major perturbations occur in the oceanic carbon reservoir, as documented in organic carbon and carbonates phases (e.g. Permian–Triassic and Triassic–Jurassic boundary, Early Toarcian, Early Aptian, Cenomanian–Turonian boundary, Palaeocene–Eocene Thermal Maximum (PETM)). All of these events, excluding the Cenomanian–Turonian boundary, record negative carbon isotope excursions, and many authors have postulated that the cause of such excursions is the massive release of continental–margin marine gas–hydrate reservoirs (clathrates). Methane has a very negative carbon isotope composition (δ13C, ca. 60%) in comparison with higher–plant and marine organic matter, and carbonate. The residence time of methane in the ocean–atmosphere reservoir is short (ca. 10 yr) and is rapidly oxidized to CO2, causing the isotopic composition of CO2 to become more negative from its assumed background value (δ13C, ca. –7%). However, to date, only the Early Toarcian, Early Aptian and PETM are well–constrained chronometric sequences that could attribute clathrate release as a viable cause to create such rapid negative δ13C excursions. Notwithstanding this, the isotopic analysis of higher–plant organic matter (e.g. charcoal, wood, leaves, pollen) has the ability to (i) record the isotopic composition of palaeoatmospheric CO2 in the geological record, (ii) correlate marine and non–marine stratigraphic successions, and (iii) confirm that oceanic carbon perturbations are not purely oceanographic in their extent and affect the entire ocean–atmosphere system. A case study from the Isle of Wight, UK, indicates that the carbon isotope composition of palaeoatmospheric CO2 during the Mid–Cretaceous had a background value of 3%, but fluctuated rapidly to more positive (ca. +0.5%) and negative values (ca. 10%) during carbon cycle perturbations (e.g. carbon burial events, carbonate platform drowning, large igneous province formation). Hence, fluctuations in the carbon isotope composition of palaeoatmospheric CO2 would compromise our use of palaeo–CO2 proxies that are dependent on constant carbon isotope ratios of CO2.