Are similarities of temperature, snow and ice cover, and (certain) marine mammals sufficient to warrant both polar regions being considered a single object of study or governance? We argue that their treatment as a unit is an invitation to examine the motivations behind the choice to be polar rather than Arctic or Antarctic. For individuals such as James Clerk Ross or Roald Amundsen, logistical requirements and analogous goals facilitated careers spanning both the Arctic and the Antarctic. This trend continued through the 20th century as individual scientists studying phenomena such as glaciers, sea ice, or aurora defined their research as “polar” in nature. Organisations such as the Scott Polar Research Institute and Norwegian Polar Institute could draw on traditions of national exploration in both polar regions, while the Arctic and Antarctic Research Institute in St. Petersburg gained its southern mandate with the importance of the International Geophysical Year. By comparison, neither the Arctic Institute in Copenhagen nor the Argentine Antarctic Institute felt any need to become polar. The creation of polar identity is ultimately a matter of geopolitics, of the value states see in instruments and symbols that speak to polar rather than Arctic or Antarctic interests. In cases such as Finland’s icebreaker industry, a technological capability justified Antarctic interest even without any national research tradition. We conclude by asking whether there is anything more natural about the polar regions than there is about the concept of a “tripolar” world in which the high alpine regions form a natural unit along with the Arctic and Antarctic.

The principle of large-scale anisotropy due to small-scale layering is applied to thermal convection. The motion takes place in a bounded porous medium heated from below. The medium is periodically layered with respect to permeability and thermal conductivity. The onset of convection as well as slightly supercritical convection are investigated. Anisotropic modelling proves useful even for small numbers of layers as long as the motion is of ‘large-scale convection’ type (Masuoka et al. 1978). The modelling always fails for motion of ‘local convection’ type.

Ultramafic xenolith lithologies representative of the mantle beneath the Midland Valley of Scotland comprise magnesian peridotite (predominantly spinel lherzolite) and cumulate wehrlites and clinopyroxenites. The lherzolites are typical of the worldwide type I (Cr-diopside) xenolith suite; their textures and mineral chemistry record a complex thermal and deformational history. The petrographical and mineralogical features of the wehrlite-clinopyroxenite suite can be interpreted within the context of a sequence of cpx + ol ± sp cumulates that have undergone a protracted period of subsoli dus re-equilibration. Mineral compositions are similar to those of type II (Al-augite) ultramafic xenolith suites.

Although xenolith populations imply widespread lower crustal heterogeneity, meta-igneous basic granulites form a major component of this region beneath the Midland Valley. They are principally composed of pi + cpx + mt ± opx ± ap; a modal continuum exists from clinopyroxenite to anorthosite, skewed towards plagioclase-rich lithologies. Garnet is rare. Locally, evidence indicates re-equilibration from garnet granulite precursors. Rock densities range from c. 2·8 to c. 3·2 gm . cm−3, implying P-wave velocities in the range 6·5–7·5 km . s−1 consistent with the known seismic properties of the lower crustal layer beneath the Midland Valley. Phase relations are also consistent with equilibration of primary assemblages at depths of 20–35 km. Retrograde reactions indicate that portions of the crust may have had a complex pressure-temperature-time evolution. Major element compositions of the granulites are broadly basaltic, ranging from ne to hy normative with the mean corresponding to alkali olivine basalt. They are distinct, chemically, from basic granulites of the Lewisian complex of NW Scotland.

Late Palaeozoic alkalic basalts in and around the Midland Valley of Scotland contain a wide variety of ‘plutonic’ xenoliths. Pyroxene-rich ultramark xenoliths (wehrlites, clinopyroxenites and garnet pyroxenites) may be representative of younger components within a dominantly peridotitic upper mantle represented by ubiquitous magnesian peridotite xenoliths. Glimmerites and other biotite-rich ultramafic xenoliths are probable samples of metasomatised upper mantle facies.

Xenoliths composed mainly of plagioclase, clinopyroxene ± orthopyroxene ± magnetite are widespread. These pyroxene granulites may typify the lower crustal layers. Garnet granulites are rare; such rocks may formerly have been important with loss of garnet occurring through retrograde metamorphism. Anorthositic xenoliths are relatively common. The lower crust may consist largely of anhydrous rocks, of gabbroic to anorthositic composition, ccurring as stratiform bodies of metacumulates.

Other xenoliths of igneous origin include tonalitic and trondhjemitic gneisses. Although these may play some role in the lower crust, they may be more abundant in the mid-crustal domains underlying the deformed upper Precambrian and lower Palaeozoic supracrustal strata. Xenoliths of quartzofeldspathic, granulitic gneisses containing garnet ± sillimanite ± rutile are also of widespread occurrence; many of these are of metasedimentary provenance and are regarded as being derived from the mid-crustal layers beneath the Southern Highlands, Midland Valley and Southern Uplands and their Irish counterparts.