Density variations and viscous flow within the convecting mantle generate deflections of the Earth’s surface with amplitudes of ±1 km and wavelengths of ~1000 km. Geoidal, shear-wave velocity, and oceanic residual depth anomalies allow present-day dynamic topography to be estimated. However, the mantle convective planform evolves through time. Direct evidence of positive dynamic topography is dicult to obtain from the stratigraphic record since elevated topography is rapidly denuded. Anomalously hot asthenosphere generates a combination of surface uplift and mantle melting. Therefore, intraplate volcanism can act as permanent record of positive dynamic topography. This study aims to determine whether the distribution and composition of intraplate volcanic rocks can be used to track the existence of anomalously hot asthenosphere, both at the present day and in the geological past. Whole-rock and isotopic analysis of volcanic samples are employed to characterise the composition of the asthenospheric mantle and isolate samples that have experienced limited lithospheric processing. Forward and inverse modelling of major and rare earth elements are used to estimate asthenospheric temperatures and lithospheric thicknesses. These geochemical results are compared with tomographic models and a variety of geological and geophysical observations are then combined to determine the degree of dynamic support. This methodology is employed in three regional studies: North Africa, Anatolia and Australia. Across North Africa, domal swells capped by Cenozoic volcanism are determined to be sub-lithospherically supported. 88 volcanic rocks from Libya and Chad were analysed using XRF and ICP-MS techniques. Asthenospheric temperatures and lithospheric thicknesses of 1300–1360 oC and ~55–75 km are estimated for volcanic provinces in both locations. The highest temperatures are predicted to exist beneath the western edge of the Sirt Basin, Libya. Anatolian topography is charaterised by high elevation, low relief plateaux. Miocene marine sedimentary deposits occur at elevations of ~1 km which indicates that this region has experienced significant youthful uplift. A transition from subduction-driven to intraplate volcanism is observed across Anatolia at ~10 Ma. Subse- quently, geochemical and tomographic temperature estimates document an east-to-west decrease in mantle temperature. This trend is corroborated by the pattern of uplift estimated by inverse modelling of longitudinal river profiles. Cenozoic magmatism of Australia is manifest both as linear chains of volcanoes and seamounts, which track the passage of mantle plumes beneath the plate, and suites of non-age-progressive volcanic fields. 76 new samples were collected and analysed from both progressive and non-progressive volcanic fields. Volcanoes that lie along hotspot tracks require asthenospheric temperatures 50–100 oC hotter than ambient mantle. Active volcanism in northern Queensland requires temperatures 30–50 oC hotter than ambient mantle and so these fields may represent the arrival of a new hotspot beneath Australia. Finally, by comparing a global database of Neogene and Quaternary intraplate volcanism with surface-wave tomographic models, it is evident that intraplate volcanism is confined to regions of thin lithosphere (< 100 km) with anomalously warm temperatures at depths of 100–200 km. The relationship between melt composition and mantle temperature can be obscured on local scales by mantle chemical heterogeneity. My global analysis reveals coherent signal between these observations and so the distribution and composition of intraplate magmatism appears to be a powerful predictor of past mantle conditions.