The Icelandic plume, a major convective upwelling, has had a considerable influence on the geological evolution of the North Atlantic region. Direct manifestations of this major convective upwelling include positive residual depth anomalies and long wavelength free-air gravity anomalies, both of which reach from Baffin Island to Norway and from Newfoundland to Svalbard. Signifi cant shear wave velocity anomalies, observed in full-waveform tomographic models between 100 km and 200 km depth, show the Icelandic plume has a complex, irregular planform. These anomalies suggest about fi ve horizontal fi ngers radiate away from the central plume conduit. The best imaged fingers lie beneath the British Isles, southern Scandinavia and Greenland, extending ~1,000 km from the Icelandic plume. It is proposed that these radial miscible fi ngers develop due to the Saffman-Taylor instability, a fluid dynamical phenomenon which occurs when a less viscous fluid is injected into a more viscous fluid. Mobility ratio (i.e. the ratio of fluid viscosities), Peclet number (i.e. the ratio of advective and diffusive transport rates) and thickness of the horizontal layer into which the fluid is injected, together control the presence of fi ngering due to the Saffman-Taylor instability. Estimates for the Icelandic plume suggest the mobility ratio is at least 15, the Peclet number is ~ 2 x 10⁴, and the asthenospheric channel thickness is 100 ± 50 km. Appropriately scaled laboratory experiments play a key role in developing a quantitative understanding of the spatial and temporal evolution of mantle plume planforms. My results prove that the presence or absence of radial miscible fi ngering due to the Saffman-Taylor instability is controlled by changes in mobility ratio, Peclet number and horizontal layer thickness. At large horizontal thicknesses, gravity has an increasingly important influence and acts to damp the production of radial viscous miscible fi ngers. Observed values from the Icelandic plume suggest the fluid dynamics may be more complex than the Saffman-Taylor instability alone. Additional processes, such as interaction with the base of the lithospheric plate, along with the Saffman-Taylor instability, may be the origin of the fi ngers.