With all due respect to basalt, and I appreciate a granite as much as the next person, kimberlite is hard to beat. But what is a kimberlite? Kimberlites2 may be classified as igneous rocks but it is difficult to know how exactly to describe them in terms of magma, at least in any conventional sense of the word. Kimberlites tap the deepest recesses of our planet that we can sample. Propelled by a formidable volatile load, kimberlite melts transit hundreds of kilometers of mantle and crust, perhaps in just a few days, to form unique ballistic deposits at Earth’s surface. Kimberlites accumulate and transport ripped-up bits from throughout most, if not all, of their ascent path, including diamond, that classic gem of desire with its remarkable qualities that have fueled a global market. Indeed, much of our understanding of kimberlite is owed to the intrepid explorers who searched for and studied elusive diamond deposits. Adding to the veil of petrological complexity kimberlites are often pervasively altered by fluids, some of which were magmatic but some of which were not. The study of kimberlites over many decades has revealed glimpses of their origins and the paths by which they have travelled.

To begin to understand kimberlites we start by considering where and when they occur. Kimberlites typically form giant carrot-shaped deposits symptomatic of their volatile-driven high-velocity emplacement, being fed at depth from a complex network of cracks and veins that draw melt out of the mantle. They are found only in continental settings (none are known from ocean basins) and seem to be intimately related to the thick lithospheric keels that anchor our landmasses into the deep mantle. Kimberlites first appear in the rock record about three billion years ago but are conspicuously concentrated in the last several hundred million years. These space-time relations tell us that kimberlite melts do not form in shallow upwelling mantle and that they are somehow related to a deep-seated geodynamic process that began during the late Archean, eerily coincident with the onset of plate tec-tonics as we understand it.

Before addressing more speculative matters, what are kimberlites made of? In general, kimberlites are something of a grotesque mixture of phenocrysts, xenocrysts, xenoliths, lithoclasts, lapilli and all manner of groundmass. Many are full of foreign materials, making the original magmas all but impossible to reconstruct. Kimberlite melts excoriate the rocks they pass through, tearing up chunks of the mantle from depths of several hundred kilometers. They may (but not always) bear diamonds, which form at depths greater than about 150 km. These crystals themselves harbor mineral and fluid inclusions that reveal a rich story of how they formed and the processes affecting their host rocks. Although most diamonds are from the cooler lithosphere, some originate at much greater depths and bear direct witness to processes in the deep, convecting upper mantle, in the transition zone, or even in the lower mantle.

Defocusing somewhat from the inherently complex detail, while eyeing those rare occasions when kimberlite magma crystallizes underground rather than exploding to the surface, we can see that kimberlite is an ultramafic rock. Kimberlite has low silica and aluminum, is high in magnesium, is low in total alkalis yet typically high in potassium relative to sodium, geochemical features that link kimberlite melts to a deep mantle source. A defining feature is that kimberlites are rich in carbon dioxide and water, which are stored in minerals such as calcite, phlogopite, and serpentine. Just how much of this vaporous hoard is primary is difficult to assess yet is fundamental to deciphering a kimberlite’s origin. Kimberlites are also inordinately well-endowed in magmaphile trace elements, enrichments that indicate an origin in low-degree partial melts of the mantle. Kimberlites carry an isotopic essence most consistent with a convecting mantle source, rather than other types of somewhat similar magmas such as their “orangeite” cousins that equilibrate with the old, cold lithosphere through which they pass.

Evidently, kimberlite melts tap the deep mantle, but where do they originate, and why? And how is one to identify, much less separate, the original melt components from the constituents that have been added during the crystallization and transport of a several hundred–kilometer vertical ride? One certainty is that carbon dioxide and water are essential ingredients in kimberlites. This is not surprising because both substances radically reduce the melting points of the rocks in which they reside; they also enter almost wholesale into the melt. A quandary is whether both volatile species were present in mutually significant amounts during original kimberlite formation. Another unknown is to what degree either were incorporated later in the melt’s evolution or post emplacement. Experiments indicate that low-degree melts of volatile-rich mantle at deep upper-mantle conditions have characteristics broadly similar to what might ostensibly be a primitive kimberlite. It seems that proto-kimberlite melts originate as some kind of a hydro-silico-carbonatitic melt generated beneath continental lithosphere (e.g., > 250 km) that becomes modified as it ingests and digests mantle rock on its way to the surface. Understanding the physical, chemical, and dynamic details of this process is where the action lies in terms of kimberlite generation.

Kimberlite magmas and their precious cargo provide a truly unique window into mantle processes and dynamics. The bits of mantle rock they sample have allowed geologists to reconstruct the lithospheric thermal structure, composition, and age. But these deep mantle melts may ultimately owe their existence to volatile elements once located at Earth’s surface and subsequently transported to depth. The diamonds that kimberlites sample from beneath the lithosphere provide a wealth of evidence for recycling of oceanic crustal materials and a key role for both carbonated and hydrated fluids in their origin. Some diamonds have mineral inclusions that are thought to have originated from the transition zone or even the lower mantle, suggesting an almost unimaginably deep kimberlite source region. An intriguing question is whether diamonds and kimberlites form as a result of the same fundamental process.

The modern style of plate tectonics, characterized by opening and closing of ocean basins, is driven by subduction of oceanic lithosphere into the mantle, a process that possibly began during the Archean or early Proterozoic. Subduction provides a convenient mechanism for transporting carbon - and water-rich fluids into the deep mantle where they are released at upper mantle to transition zone depths in a series of dehydration and decarbonation reactions. Liberation of these fluids beneath stable continental lithosphere effectively creates a continuous mechanism for “juicing up” the sublithospheric mantle over time. These volatile-charged, inviscid and buoyant fluids could be truly proto-kimberlitic, the original agents of chemical mass transfer that stain the mantle with slab components. Perhaps such volatile-charged mantle is always poised to melt to a low-degree given the right tectonic impetus, with melts accumulating beneath the lithosphere only to “pop up” every now and again to provide yet another exquisite probe into the deep mantle. Kimberlites and their putative origins are indeed bewildering. But they reveal more about Earth’s deep interior than any other rock or magma type.

Kimberlite, as used here, refers to Group I or ‘archetypal’ kimberlite but not rocks formerly called Group II kimberlite or orangeites that are now best referred to as types of lamproite, derived from partial melting of enriched lithospheric mantle sources (see Pearson et al. this issue, page 387)
The readers has free access to the “free” material but MSA holds the rights