Journal of Petrology Volume 41 Number 7 Pages 897-899 2000
© Oxford University Press 2000
A Collection of Papers to Commemorate the Work of Keith Gordon Cox, 1933–1998
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION REFERENCES |
|---|
On a blustery wet Saturday afternoon in October 1998, about 250 people met in the University Museum in Oxford to commemorate the life and work of Keith Cox, who had drowned in a sailing accident off the coast of Mull just over 2 months before, on 27 August. The feeling of shock was still acute and the sense of loss still tangible: Keith had been such an influence, in one way or another, on everyone there. Many were students whom Keith had taught, supervised or examined; others were colleagues with whom Keith had agreed, or argued; others were old friends from his student or army days; and his family was there, too, including his wife Gillian, who had survived the accident and tried bravely and so hard to save him. Several of those present have contributed papers to this specially dedicated Issue of the Journal of Petrology, an entirely fitting tribute, not only to one who guided the Journal as Managing Editor, then Editor, over 12 years between 1972 and 1984, but to one of the most meticulous and influential petrologists of his generation.
Influence works both ways. Probably the reason why Keith considered geology in the first place was that his father, Sir Gordon Cox, appointed Professor of Inorganic and Physical Chemistry at Leeds University when Keith was about 15, was a faculty friend of W. Q. Kennedy (reviver and extender of the use of the term tholeiite), himself later to be such a shrewd guide to Keith when he joined the Research Institute of African Geology at Leeds University as an Oppenheimer Scholar (Ph.D. research student) in 1956.
But Oxford was to come first and Keith went up to the Queen’s College in 1952, having completed his National Service as a lieutenant in the Royal Engineers (by whom, incidentally, particularly by ex-Wermacht soldiers attached to his squadron in Berlin, he was so well instructed in the skills of working large earth-moving equipment that he subsequently felt able to challenge any other Fellow of the Royal Society to competition in the use of the bulldozer). At Oxford the Head of Department was Lawrence Wager, founder of this Journal in 1960 with fellow editors, T. F. W. Barth, C. E. Tilley and H. S. Yoder. It is particularly appropriate that one of the authors in this Memorial Issue of the Journal, Peter Wyllie, not only contributed a paper to its first issue (Wyllie, P. J. & Tuttle, O. F., The system CaO–CO2–H2O and the origin of carbonatites) but was co-author, with G. M. Biggar, of Keith’s first paper in the Journal in 1962 (The habit of apatite in synthetic systems and rocks). That first issue also contained a paper on a subject, basalts, in which Keith himself was to achieve, perhaps, his greatest distinction (the paper was H. Kuno, High-alumina basalt). Inevitably there was also a paper on cumulates (L. R. Wager, G. M. Brown & W. J. Wadsworth, Types of igneous cumulate). Wager taught his students the importance of fractional crystallization as a mechanism in the formation of igneous rocks and Keith remained firmly of the same opinion throughout his career. Another influence at Oxford was Keith’s own tutor, J. V. Harrison, an austere and often kilted Scot who did not tolerate fools gladly. Keith had something of that about him. Harrison, famous for his pioneering work, especially on gravity tectonics, with the then Anglo-Persian Oil Company in Iran, fostered Keith’s love of fieldwork and geological mapping. The quality of Keith’s fieldwork and draughting skill can be seen in many of his later publications, perhaps particularly those incorporating the research done while he was at Leeds on the Nuanetsi Igneous Province. In view of Keith’s later reputation it is worth remembering that at this time his research work was directed by W. Q. Kennedy as part of the latter’s grand design to study the ‘Younger Granites’ of Africa, and late in his first field season he and C. J. Stillman were ordered to Nyasaland to map the Chikala Hill Complex, a syenite and nepheline syenite body. This illustrates some of the range of Keith’s experience and expertise but, naturally, the most fruitful result of this period was his first acquaintance with the Karoo flood basalts and the connection of such phenomena with large-scale tectonism.
|
Keith’s move to the Grant Institute of Geology at Edinburgh University in 1962 took him to a department favourably equipped and opportunely staffed to welcome someone so keen to expand research on basalts and their origins, and numerous publications duly emerged. On the grand scale the Karoo figured most prominently, with the first distillation of Keith’s ideas relating igneous and tectonic processes eventually appearing in The Karoo Volcanic Cycle in the Journal of the Geological Society of London in 1972. In this paper Keith advanced a model that traced the evolution of a mantle-melting event and its consequences from its early fractionated products and succeeding highly primitive picritic representatives, through a subsequent, lengthy ‘steady-state’ phase of relatively uniform, moderately fractionated basalts, into a waning phase of mainly alkaline rocks. It was further suggested that ponded potassium-rich picritic magma might offer a source for the southern African kimberlites and that acid volcanics such as the rhyolites of the Lebombo Monocline might be the anatectic results of magmatic heating of the crust. (One irreverent Oxford colleague suggested that, because of this dual magma production process, the title of the paper should have been The Karoo Volcanic Bicycle.) Finally, a sign of things to come, Keith predicted that Karoo-type patterns of igneous activity might be discovered in other flood-basalt provinces such as the Deccan and Greenland. He had observed the tectonic similarities between the Lebombo Monocline and the East Greenland coastal monocline originally discovered by Wager.
Keith never got to Greenland but he did extend his researches on flood basalts to the Deccan while he was still at Edinburgh, and then to the Parana and its detached fragment, the Etendeka, and the Scottish Hebridean province after his return to Oxford. The contributions in this Issue by Krishnamurthy et al., Mahoney et al., Widdowson et al., Ellam & Stuart and Scarrow et al., of whom several were Keith’s research students, specifically show that he had identified some ground that has continued to be very fertile. He himself was prepared to till that ground with whatever tool promised the best results. In the case of the Deccan he saw at the start that there was a fieldwork problem in establishing the stratigraphy that would have to be solved if the petrological researches were to have reliable chronological calibration. Characteristically, he first went out with an alidade and a clinometer to do some precise mapping: the regional dip proved in some areas to be 0·5°. Trace element and isotope geochemistry were subsequently employed for the same ends as well as for identifying mantle reservoirs or crustal involvement in magma evolution.
Keith’s credo for flood basalts is embodied in the seminal paper A model for flood basalt vulcanism published, appropriately, in this Journal in 1980. Sparse picrites could be erupted early if the crust were sufficiently thinned and could be primary but the bulk of flood basalts had Fe/Mg ratios inappropriate for primary mantle melts: they were processed by polybaric fractional crystallization and their major element uniformity could be explained by the buffering effect of the separation of olivine, clinopyroxene and plagioclase. For reasons of density, much of the magma formed in the mantle ponded at the crust–mantle boundary forming sill complexes that both staged the upward transfer of fractionated liquids and in time profoundly changed the crustal structure by massive underplating. Characteristically simple and elegant in its exposition, the model has profoundly influenced virtually all subsequent thinking about the origin of flood basalts, perhaps not excluding their possible oceanic equivalents, deep ocean plateaux, as the papers by Kamber & Collerson, Révillon et al. and Kerr et al. in this volume testify, as well as transitions from continental to oceanic settings (Fitton et al.) and more ancient sub-continental mantle regions (Menzies et al.).
On a smaller scale in terms of the volumes of rock concerned, the Aden Volcanics provided evidence from the work of Keith and his colleagues and students that strengthened his conviction that fractional crystallization had to be a mechanism of major importance in determining the evolution of igneous rocks. Field evidence, petrography, and major and trace element chemistry were all correlated in a simple model. The flat-lying lavas of the Shamsan Caldera sequence of the Aden Volcano with their upward decrease in silica and alkalis and increase in phenocryst content could be satisfactorily explained as the products of a progressive emptying of a fractionated magma chamber whose lower levels had been progressively enriched in settled crystals. Everything seemed to fit into place. Well, not quite everything: Sr-isotope data from the Shamsan lavas proved to have too great a range for a uniform parent magma. Thenceforth Keith invariably included the isotopic dimension (Braun et al. in this Issue demonstrate some of the latest developments in this approach) in his petrogenetic thinking, and further work on the Deccan lavas published in this Journal saw not only collaboration with one other author in this Issue, C. J. Hawkesworth, but support for the innovative suggestions of another, M. J. O’Hara. The paper concerned, Cox & Hawkesworth (1985), concluded that major element variation in the Deccan lavas, controlled by gabbro fractionation, was, as shown by Sr-isotope data, modified by crustal contamination, particularly evident in the lower part of the succession and decreasing upwards; there was also evidence that some of the lavas were derived from enriched mantle comparable with the source of oceanic basalts, a topic addressed by Kamber & Collerson and Garcia et al. in this Issue. Cox & Hawkesworth pointed out that their results provided support for the model of RTF (replenished–tapped–fractionating) magma chamber evolution developed especially by M. J. O’Hara, whose contribution to this Issue widens the field of scrutiny in a stimulating way.
More than two-thirds of Keith’s papers were published after his return to Oxford in 1972, the year in which he became a Managing Editor of this Journal. Many were concerned with advances in research areas first explored while he was at Leeds and Edinburgh, including his contributions on potassium-rich rocks of the Roman Region and kimberlites. The ultramafic xenoliths in kimberlites in particular drew his attention as an acute petrographer interested in textures, as well as one needing to know about mantle composition and mineralogy.
New research areas added to the list were the Falkland Islands and the Tertiary plateau lavas of Skye in the Scottish Hebrides. The latter furthered understanding of plume-related flood-basalt genesis, a subject re-examined in this Issue by Ellam & Stuart and Scarrow et al. The former traced by palaeomagnetism the migration of another piece in the jigsaw of Gondwanaland fragmentation. The particular poignancy of the Falklands interest is that a study of the petrology, petrogenesis and geochemical provinciality of the dolerite dykes linking them with magma types of the Karoo and the Ferrar province in Antarctica was Keith’s last, collaborative, publication and that, in 1999, it was published posthumously.
Although Keith’s most productive period was the 1980s, when he averaged more than two papers a year, he was still very active at the time of his death, with some important work on the stocks and ideas forming in his mind for a general textbook on igneous petrology. Given the success of the other two textbooks, of which he was senior author, one can imagine how enlightening and inspirational that might have been. His dedication to this Journal was keen, critical and constant. It is the repository of more of his papers than any other journal and so it is an important part of his legacy to the science of petrology in particular, and earth sciences in general. It is a rich legacy, which will continue to fund enquiry for as far ahead as we can see.
David Bell
Department of Earth Sciences, Oxford University
10 March 2000
|
REFERENCES |
|---|
|
TOPINTRODUCTIONREFERENCES |
|---|
Kuno, H. (1960). High-alumina basalt. Journal of Petrology 1, 121–145.
Wager, L. R., Brown, G. M. & Wadsworth, W. J. (1960). Types of igneous cumulates. Journal of Petrology 1, 73–85.
Wyllie, P. J. & Tuttle, O. F. (1960). The system CaO–CO2–H2O and the origin of carbonatites. Journal of Petrology 1, 1–46.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||