Appendices for: John Stuart Webb, FREng, and Applied Geochemistry at the Imperial College of Science and Technology, London

ABSTRACT The life and work of the pioneering applied geochemist, Professor John Stuart Webb (1920–2007), FREng, founder and long-time Director of the Geochemical Prospecting Research Centre (1954–63) and Applied Geochemistry Research Group (1963–79), and Senior Research Fellow until 1988 is reviewed. The work began with his recognition of geochemical provinces as the key to location of areas which might contain mineral deposits, initially recognized in New Brunswick, proven in Zambia and Sierra Leone. The initial focus of the work was on mineral exploration but, with the passage of time, it broadened to embrace: multi-purpose geochemical mapping; agricultural and environmental geochemistry; the publishing of pioneering regional geochemical atlases of Zambia (Webb et al. 1964a), Derbyshire (Nichol et al. 1970b), Denbighshire (Nichol et al. 1970b), Devon and North Cornwall (Nichol et al. 1971), Northern Ireland (Webb et al. 1973) and England and Wales (Webb et al. 1978); applied marine geochemistry (especially the investigation of metalliferous brines and manganese nodules); and urban geochemistry. Over 100 students have now graduated with higher degrees from the school which Webb began, to apply the techniques in industry, geological surveys, or to train a new generation of exploration and environmental geochemists around the world. Sadly, the recognition which he well deserved was much greater abroad than in his home country. Following his retirement, research in environmental and marine geochemistry continued to prosper for another 20 years. Supplementary material: A complete list of all GPRC/AGRG publications, plus the few AGRG ‘Technical Communications’ not concerned with analytical methods, is given in Supplementary Appendix 1; of Technical Communications on analytical methods in Supplementary Appendix 2; of GPRC/AGRG theses in Supplementary Appendix 3; and of technical personnel in Supplementary Appendix 4. All are available at http://www.geolsoc.org.uk/SUP18422.


SUPPLEMENTARY MATERIAL:
A complete list of all GPRC/AGRG publications, plus the few AGRG 'Technical Communications' not concerned with analytical methods, is given in Supplementary Appendix 1; of Technical Communications on analytical methods in Supplementary Appendix 2; of GPRC/AGRG theses in Supplementary Appendix 3; and of technical personnel in Supplementary Appendix 4. All are available at http://www.geolsoc.org.uk/SUP18422. The death of Professor John Stuart Webb, FREng, Emeritus Professor of Applied Geochemistry at the Imperial College of Science and Technology, London, on 2nd April 2007, marked the end of an era in British, as well as international, geochemistry. One of his former research students, who first knew him as a geochemical consultant in Africa in the 1960s, recalled him, 35 years later, as: an excellent leader and teacher, someone who understood him both as a person and as a research student, focussing his research to match his personal experiences and scientific abilities; strong in the fundamentals of geology; a man who could quickly assess a field geological problem, which would allow him to focus on the critical geochemical parameters necessary to solve that problem; and someone who never deviated from the basic fundamentals of science, so that his Research and Development had to have a sound scientific base and the highest quality control possible, ensuring that the results were real and not just potentially of statistical interest. "The research studies, therefore, have been the basis of the development of applied geochemistry and the results will stand the test of time." (pers. comm., J.L. Walker to I. Nichol 1999).
Together with the Australian-born Canadian geologist and pioneer of biogeochemistry, Harry Varney Warren , American geologist and geochemist, Herbert Edwin Hawkes , and Canadian mining geologist and geochemist, John Evans Riddell , Webb is rightly regarded as one of the 'founding fathers' of applied geochemistry in the West and, in Britain, as 'the pioneer of applied geochemistry in this country, he helped to show that the subject was as important as geophysics ' (S.H.U. Bowie, pers. comm. 1999). Webb was dubbed the 'Father of English geochemical mapping' by Professor Edward Howel Francis on the occasion of the Geological Society of London's award of The William Smith Medal to Webb in 1981(Francis 1982. Over the years in which Webb was the Director of the Geochemical Prospecting Research Centre  and its successor, the Applied Geochemistry Research Group (1963-79, when he retired to become Senior Research Fellow until 1988, in the Royal School of Mines (RSM), Imperial College of Science and Technology (ICST), London, he had the vision to broaden the scope of its work from an initial focus on exploration for concealed orebodies into: a) marine geochemistry, concerned with the exploration for marine mineral deposits; b) regional geochemistry, concerned with the geochemical mapping of large land areas as an aid to the delineation of geochemical provinces (areas of the Earth's crust in which the chemical composition is significantly different from the average); c) agricultural geochemistry, concerned with the geological sources and health effects of the major and minor elements necessary in plant and animal nutrition; d) environmental geochemistry, concerned with the pollution of the surficial environment by industrial waste; e) geochemistry in public health, concerned with the minor elements in foods and drinking water and their relation to human health; f) interpretation of the geology, as well as the recognition of areas in which minor element deficiencies or excesses might warrant more detailed investigation as a result of their potential relevance to applications (c), (d) or (e) (above). The definitions used above are those of Rose, Hawkes & Webb (1979).

War service
Following a brief period as assistant mining geologist to the Government's Non-Ferrous Metallic Ores Committee, Webb joined the Royal Engineers in 1941 ( Fig. 1), having volunteered for their tunnelling battalion. He became a sapper, and was subsequently promoted to Lance Corporal. By 1943, there was an urgent need to look for additional sources of the metal tantalum. It is a superhard, refractory, metal, soluble in acids only with great difficulty, and is easily fabricated. First used as a filament in incandescent lamps in the early 1900s (Miller 1959), by the early 1940s its usage had spread to machine tools, radar, and to use in nuclear weapons and related research under the Manhattan Project . At some point in 1943, Webb was commissioned 2nd Lieutenant at midnight, transferred to the unemployed officers' pool and, shortly after midnight, was seconded to the Geological Survey of Nigeria, as an economic mineralogist (J.S. Webb, pers. comm. 1998;J.S. Tooms, pers. comm. 2007). His job was to look for additional sources of tantalite, the ore from which tantalum is derived, in the pegmatites of central Nigeria. These were known to contain high-grade ores of this type. He worked both in the field and in the survey laboratory in Kaduna, where he undertook chemical analysis (Webb 1943). Reginald Ronald Eric Jacobson (b. 1912) a former postgraduate student at RSM in 1944-45 and1946-47 (who would soon become Director of the Geological Survey of Nigeria) and Webb showed that the bulk of the tin and niobium-tantalum ores were formed during late-stage albitization of the pegmatites, and that the pegmatites were genetically related to the Older Granites (Jacobson & Webb 1946). They also discovered, in the tin-bearing pegmatites of Kabba Province, central Nigeria, a new tin mineral, to which they gave the name 'nigerite' (Jacobson & Webb 1947), now recognized (Armbruster 2002) as a mineral group which is part of a högbomite-nigerite-taaffeite polysomatic series (see Veblen 1991, for discussion) composed of spinel and nolanite modules. Unfortunately, Webb was invalided back to England in 1944 in a poor state of health: having been unlucky enough to fall a victim to tick-typhus, he was in hospital for 3 months and was transferred to Lagos, but caught malaria on the way (he may also, have suffered from blackwater fever, a serious complication of malaria). Declared unfit to travel, he was flown home by Dakota aircraft via Dacca [Dakar] and Lisbon and, on arrival at Northolt airport, London, he could barely walk, even with the aid of walking-sticks, and was driven to his parents home. So changed was he by the effects of illness, that on arrival, the family friend who answered the door did not recognize him! However, one good thing which resulted from all this was that, while in Kaduna, he had become a pen-friend of Jean Millicent Dyer (1920-97), a friend of his sister and an accomplished pianist, who was then working as a secretary at the publishers the National Trade Press Ltd, in London. Following his recovery, they married in January 1946, both aged 25. Jean was an incredibly important and supportive person in his life. Their only son, Stuart, now a distinguished doctor, was born in 1950 (J.S. Webb, pers. comm. 1998Webb, pers. comm. , 2007Webb, pers. comm. , 2008. When not thinking about geochemistry (which seemed, sometimes, never to stop: he might phone one at home at any hour of the evening, often more than once, to discuss a new idea), he much enjoyed painting, shooting, fishing (Fig. 2), wildlife photography, and amateur radio (Fig. 3). He used to take his rods with him when travelling and is reputed to have caught a previously unknown species of fish in an African lake.
The rise of geochemical prospecting Webb was awarded a Beit Scientific Research Fellowship at ICST in 1945, and began his doctoral thesis on 'The origin and mineral paragenesis of the tin lodes of Cornwall', for which he was awarded the Judd Prize in 1946, and his PhD (Mining Geology) degree and the Diploma of Imperial College in 1947. Two publications arose from this work (Webb 1946(Webb , 1947, describing the mechanism of quartz replacement-veining in the tin-mineralised Carn Brea Granite at the South Crofty mine, near Redruth, Cornwall. After his appointment by London University as a Lecturer in Mining Geology at ICST in 1947, his research initially remained along the conventional lines of mining geology. With the cessation of the WW II hostilities, the numbers of postgraduate students in Mining Geology had begun to rise by 1948 (in part as a result of the increased post-war undergraduate intake in 1946). These included a Canadian student, Anthony R. Barringer (1925Barringer ( -2009. Following wartime service in the British Army, Barringer studied for his BSc in Mining Geology at the RSM (1948-51); and then for his PhD, 'A study of the mineralization in the pyritic deposits of Southern Spain and Portugal' (1951-54), advised by Webb and David Williams. Williams had studied the geology of the Rio Tinto Mines, Spain, in the 1930s, and had long been a consultant to the Rio Tinto Company (Williams 1963, p. 87). Work on pyritic mineral deposits was continued intermittently by Webb from 1949 onwards, aided by several final-year undergraduate studies on the geology of a number of pyrite deposits in Cyprus and Portugal together with Barringer's work. All this was summarized in a paper on the San Domingos pyrite deposit, Portugal (Webb 1958a). Barringer would go on to become a major developer of geophysical and geochemical instrumentation, and a Visiting Professor to the Applied Geochemistry Research Group (AGRG) from 1965-1981. Meanwhile, in America, following the end of WW II, the Chief Geologist of the U.S. Geological Survey (USGS) asked his staff for ideas on possible areas of new research. As a result of growing American interest in the application of geochemistry to the search for ores (Figs 4,5), the chemist and geologist, Lyman Coleman Huff  suggested the use of trace elements in surface and ground waters as prospecting aids; Helen Leighton Cannon  suggested the use of geobotanical and biogeochemical methods; and Hawkes, together with his old supervisor from the Massachusetts Institute of Technology, where he had studied for his PhD (awarded in 1940), Professor Walter Harry Newhouse (1897-1969, suggested the use of soil surveys. After consideration by senior USGS staff, all three proposals were accepted and resulted, in July 1946, in the establishment of the USGS Geochemical Prospecting Unit [later, Geochemical Prospecting Section (GPS)]. Its initial staff consisted of Huff and Cannon, with Hawkes as its Acting Chief. They were later joined by Hubert William Lakin  and others (Hawkes 1976).
In 1948, Webb was awarded the Daniel Pidgeon Fund of The Geological Society, London, to investigate tin-tungsten mineralization in the Hercynian of Western Europe. However, by this time, he too was aware of publications on geochemical prospecting using trace elements in: (a) natural waters (Vogt & Rosenqvist 1942, in Norway;Sergeev 1946, in Russia;Rankama 1947 in Finland;and Huff 1948, in the USA); and (b) vegetation (Rankama 1940Vogt et al. 1943, in Norway;Hedström & Nordström 1945, in Sweden;Maliuga 1947, in Russia;Robinson et al. 1947, in the USA;Warren & Howatson 1947and Warren & Delavault 1948, 1949and Lundberg 1948, in England); and he was evidently also in contact with the members of the Geochemical Prospecting Section of the USGS (Webb & Millman 1950. His initial interest in the possibility of geochemical prospecting was probably inspired by reading one of Rankama's papers in 1947(Webb 1982. He decided to begin research into the potential of geochemical prospecting in Africa (Fig.  6), as this would provide favourable conditions of rockweathering, vegetation and climate, as well as relatively undeveloped mineralized areas, and would also show whether methods which seemed promising in a broadly temperate setting would also work in the tropics. Very conveniently, Anthony Philip Millman (b. 1923, fl. 1971) had joined the Mining Geology Department at the ICST as an Assistant Lecturer in 1948. Millman had been an undergraduate there  on either side of his war service with the Royal Engineers in the Far East, and he had also studied some botany and zoology in addition to geology. The subject for his doctoral thesis, suggested by Webb (Millman 1953, p. 2), was 'The theory and practice of geochemical prospecting'. The initial emphasis of the study (Millman 1953, p. 105) was to be on testing the biogeochemical method of prospecting, which had been advocated by Rankama (1940). How well would methods developed in essentially temperate climates transfer successfully to the tropics? The analytical methods to be used were colorimetric tests with the reagent dithizone (an organic compound which gives a deep green solution when dissolved in carbon tetrachloride which, in the presence of lead, zinc or copper, rapidly turns first purple, then red, as their concentration increases) after Sandell (1944) andHuff (1948), for the water samples. Also employed was semiquantitative optical spectrography, using a model E492 Large Quartz Littrow Spectrograph, made by Hilger & Watts Ltd., London (Mitchell 1948), a technique which he had proved to his department head, Read, was worthwhile (when Read provided the funds to purchase the spectrograph, he remarked 'Just think how many geological hammers you could have bought with the money!'; H.H. Read quoted in Webb 1982). The spectrograph was installed with the aid of Wallace Spencer Pitcher , later to become a world leader in Read's old field, granite research, and Professor of Earth Sciences at  the University of Liverpool (Hutton 2004), who was at that time an Assistant Lecturer  in Mining Geology.
Probably with advice from Jacobson, who was, by 1948, Acting-Director of the Geological Survey of Nigeria (GSN), and with funding from the Hilary Bauerman Bequest Fund of RSM and the Mines Development Syndicate (West Africa) Ltd., it was decided to carry out a geochemical survey of the distribution of heavy metals in the natural waters and vegetation in: (1) the vicinity of the Ameka and Ameri lead-zinc lodes, located 10 miles (16 km) south of the town of Abakaliki (the capital city of Ebonyi state), in the Nyeba [Enyigba] lead-zinc mining district of southeastern Nigeria; and (2) in the vicinity of the Tunga and Farin Kaya lodes in the Wase district, lying c. 100 miles (160 km) south-east of Jos in north-central Nigeria. Both areas were in fairly flatlying savannah country.
Apparently, a soil survey at Ameka and Ameri had been planned by the GSN in 1947, at the suggestion of the then director, Dr. C. Raeburn, but it had never got beyond preliminary investigations in their laboratory at Kaduna (R.A. Mackay in discussion of Webb & Millman 1950, p. 339); subsequently R.O. Roberts analysed a suite of 26 soils, 4 rocks, 5 water and 2 leaf samples, collected by the Mines Development Syndicate (West Africa) in March 1949, over ground already covered by a GSN gravity survey. A second suite of 76 soils and 5 rocks were collected in May 1949. Roberts also used dithizone (presumably after Sandell 1944and/or Huff 1948no details are given in Roberts 1953). Webb assisted Millman to collect vegetation samples in the field in February and July 1949, using the 'grid sampling' approach of Rankama (1940). Control samples of vegetation from an area known to be negative for lead and zinc mineralization were collected for them (Millman 1953) by Dean Arthur Oliver Morgan , who had just completed his PhD in Mining Geology at RSM, and who was now working for Gold Fields Rhodesian Development Cobalt Ltd. in Southern Rhodesia as an assistant geologist. Webb and Millman also collected water samples, under various conditions of rainfall or dryness, during the wet season in July 1949. It was found necessary to modify Huff's (1948) analytical method in order to get it to work under tropical conditions. An opportunity arose, however, for testing a limited range of soil samples collected by Webb from the Wase area. The results of these were sufficiently interesting to encourage a limited orientation study to establish satisfactory analytical techniques, and for providing data on the background concentrations of the ore-metals together with anomaly widths and peak values' (Millman 1953, pp. 105-106). The initial Nigerian results were published in Millman (1950) and Webb & Millman (1950. Millman was awarded the William Frecheville  Prize of the Institution of Mining and Metallurgy (IMM), London, for his geobotanical work.
For comparison, a second investigation was next carried out by Millman in Cornwall, England, taking plant and soil samples over the tin-copper lodes of the Hingston Down Consoles Mine, West Drakewells Mine and Russell United Mine, all located in the Hingston Downes-Gunnislake area.
At this time, Hawkes too was curious to know whether the geochemical methods which the USGS were finding promising when applied to exploration problems in the USA and Canada, would work as well under tropical conditions. As it happened, the Mutual Security Agency (MSA) of the US Government was already engaged jointly with the Nigerian Geological Survey in a geological research program to investigate the lead-zinc deposits in the Nyeba area (Bogue & Reynolds 1951) and it 'could accommodate an experimental geochemical program' (Hawkes 1954, p. 52). Accordingly, he was 'on special loan' by the USGS to the MSA programme from November 1951 to January 1952, taking with him a field laboratory, assembled and packed by USGS GPS analytical chemist, Harold Bloom (1913-2001. Using colorimetric analysis following a nitric acid attack (Bloom & Crowe 1953) produced only moderate lead and zinc anomalies from soil samples taken over the Ameri lode. However, Hawkes was able to confirm the results of a vegetation traverse reported by Webb & Millman (1951). He also collected water samples in the vicinity of the Ameri lode, in an attempt to duplicate the results of Webb & Millman (1950), but it was the beginning of the dry season, and the presence of organic impurities in the water rendered the same analytical method (Huff 1948) as had been used by Webb and Millman useless. Consequently, the water sampling programme was abandoned (Hawkes 1954, p. 94).
On his return journey to America, Hawkes passed through London, and gave Webb a set of 18 samples, the <80-mesh fraction (J.S. Tooms, pers. comm. 2007;Hawkes 1954, p. 71) of residual and alluvial soils which he had collected on a traverse across the Rikaya lode at Wase. In order to establish the best analytical method, comparative tests of this suite of samples were undertaken by: (a) spectrography by Millman at the RSM (which included preparing a new base material for the analytical standards which was better suited to match the West African soils); (b) chemical assays by the Metallurgy Department at RSM; and (c) the dithizone colorimetric method, using two different extractions, at the USGS GPS. Hawkes was convinced that there was no hope of geochemistry working in areas of such thick lateritic soil cover, such as Northern Rhodesia. However, Webb showed in 1953 that if one analysed the tropical soils following a bisulphate attack, then excellent anomalies were obtained (J.S. Tooms, pers. comm. 2007). Consequently, Millman's results with the soil samples showed very good anomalies for lead and zinc over the Tunga and Farin Kaya lodes at Wase, as well as for tin over the Hingston Down North and Central lodes in Cornwall (Millman 1953, Figs 14, 17, 18). Millman (1983) concluded that, if taken at the right season, water samples could work well in regional surveys if they were collected at stream confluences, working upstream, using aerial photographs rather than topographic maps for control, as the former were more readily available. His work showed, for the first time, that the tropical savannah was suitable for vegetation-based prospecting, but that it was necessary to collect composite samples of leaves and twigs to overcome their great natural variation. Anomalous lead concentrations were found over lodes in Nigeria, and tin, copper, zinc and silver in Cornwall; but, in Cornwall, the vegetation sampling was found to be quite unsuitable, because of the sparseness of the vegetation and narrow anomalies (Millman 1957). Overall, it was concluded that soil-sampling looked to be the most promising method (Millman 1953, pp. 117, 119-121, 133-134). According to Williams (1963, p. 132), Millman also successfully located a near-surface anomaly over a lead-zinc vein in limestone concealed beneath 700 ft (211 m) of sandstone and shale, but that work is not in his PhD thesis. Nevertheless, Millman decided to remain a mining geologist, and subsequently devoted himself to ore microscopy.

North America beckons
Following their initial contact, Hawkes invited Webb to visit establishments engaged in geochemical exploration in North America, under the auspices of the MSA, as a Technical Assistance Project sponsored by the British Department of Scientific and Industrial Research (DSIR). Webb spent six weeks in the USA and Canada in early 1952 (Hawkes 1976, p. 6), 'personally conducted' by Hawkes throughout. Together they saw work, in both field and laboratory settings, with: the American Smelting and Refining Company (Tennessee and elsewhere); American Zinc Company (Tennessee); Consolidated Mining and Smelting Company of Canada (British Columbia); Geochemical Surveys (Texas); Hecla Mining Company (Coeur d'Alene); Kennco Explorations (British Columbia); New Jersey Zinc Company (Virginia and Tennessee); Research Incorporated (Texas); the US Geological Survey laboratories in Denver and Washington, DC, and USGS geochemical field parties at Cobalt and Coeur d'Alene (Idaho) and on the Colorado Plateau. They also visited Warren to discuss his biogeochemical research at the University of British Columbia (Webb 1953, pp. 322-323).
In the course of his tour, Webb saw current work on rocks, residual and non-residual soils, stream waters and stream sediment, and plants. Gaining knowledge of the analytical techniques being developed was equally important, for although Webb, Williams, and their students, had been experimenting with spectrographic methods for determination of trace elements in ore minerals and other materials since 1947 (El S. Mohamed El Shazley, PhD 1951;Kallat Venugopal, PhD 1952;and Millman 1953), the equipment was unsuited to both rapid throughput of samples and use in the field and it was obvious that more research was needed on rapid chemical methods suitable for field use under both temperate and tropical conditions. Despite early work in Russia at a mining-camp scale (summarized in Garrett et al. 2009), searching for bedrock tin deposits by 'stannometric' surveys (Flerov 1935), and an American suggestion (Campbell 1941) that spectrographic analyses of placer gravels (e.g. for tin), sampling systematically upstream, might locate 'the mother lode', at the time of Webb's visit, the idea of sampling active drainage sediments was novel. It had only just been tentatively tested in the USA by geologist, Thomas Seaward Lovering (1896-1991) of the USGS Mineral Deposits Branch, together with Huff, and GPS chemist, Hy Almond , in the course of examination of a virgin porphyry-copper deposit in Arizona, using a mobile wetchemical laboratory mounted in the back of a small van (Lovering et al. 1950, fig. 3).
Because of the desert setting of this deposit, little copper from the orebody dissolved in groundwater or runoff, nor was it taken up by plants growing on the small outcrop (although some possible 'indicator' plants were found). However, mechanical disintegration of the copper mineral chrysocolla, present in joints running through the orebody, during weathering carried it into the fine fraction of the soil. The copper-rich soil then migrated downslope, by slope-wash and soil-creep, until it was carried into the nearby drainage channels, where it became incorporated into the alluvium and swept downstream during floods. They found that the fine fraction of the alluvium still showed a 'significant copper anomaly' 1.4 miles (2.2 km) downstream of the outcrop of the deposit (Lovering et al. 1950, pp. 512-513).
By 1952, USGS staff were investigating the systematic upstream sampling of confluences in the main waters of a drainage system, following anomalous concentrations of copper, lead or zinc to the point at which the metal appeared to be entering the surface waters (Webb 1953, p. 333;Millman 1953, pp. 117-118). Although the investigations were still underway at the time of Webb and Hawke's visit, 'authoritative opinion . . . holds that under favourable conditions a geochemical survey of a drainage system could provide valuable information regarding the possible existence of mineralization within a catchment area, and might, therefore, prove a powerful tool in regional reconnaissance' (Webb 1953, p. 334).
On his return to England in 1952, Webb wrote up a detailed report, in which he concluded that there was great scope for application of the various methods, both in Britain and elsewhere in the British Commonwealth. 'Three lines of attack present themselves: (a) the critical investigation of known geochemical methods under a variety of new conditions; (b) the development of new techniques to meet different demands; and (c) research on the fundamental principles and processes involved (in the formation and detection of geochemical dispersion and anomalies). It is considered that the most effective course to follow would be a programme of research and development organized on the same general pattern as that which has proved successful in America. This would entail provision for closely co-ordinated field investigations, routine analysis and analytical research . . . Given adequate arrangements, arrangements might be made to hold in London a short, intensive course of general instruction on geochemical methods. Such a course would constitute an effective first step in the broad programme of investigation outlined above' (Webb 1953, p. 345).
One day in 1952, Keith Alan Viewing (later to take his PhD in the GPRC in 1963) was working as a geologist with Rhodesia Selection Trust (a company formed by Sir Alfred Chester Beatty (Snr) to exploit the copper deposits) mapping the Lower Roan Formation of the Katanga, looking for new copper deposits. Having loaded up the team's lorry at the great Mufulira Copper Mine, Zambia, ready to return to the bush on the Lukanga River, several hours drive away 'we turned into the huge core yard, normally at peace, but on this day the tiny laboratory, a small room with a rough wooden bench and a sink, in the centre was crammed. They were packed inside that corrugated shed and crowded around the door and only window. Silence. Not much room, but you could see a small quiet man with the most immense brown eyes that twinkled as he talked. What did he say? Difficult to hear, and as his tale unfolded attention was lost . . . Then there was a demonstration; two fists, each containing a dark green liquid; a vigorous shake, to obtain the emulsion, then a long beat as he shook the two handfuls of test tubes back and forth from chest to shoulder, and the only sound was the swish of bubbles in the tubes and the gasp of breath from John Webb. The magic continued for two minutes exactly. The test-tubes contained a green organic reagent, a buffer to control the pH, and a sample of soil. There were all shades from green to pink as we saw for the first time, a simple chemical analysis of copper in the range from say 25 to 500 ppm. This was magic indeed, a practical and cheap field method to reach beneath the overburden and indicate hidden deposits in the weathered rock, say 15 m below . . . It made a change from looking for copperthiocyanate crystals under the microscope. . . . ' (pers. comms., to I. Nichol 1999;RJH 2007).
While Webb had been abroad, Williams recruited John Somerville Tooms (b. 1927) as a research student. Tooms had just graduated from the University of Wales with a first class Honours BSc in geology, following his wartime service in the Parachute Regiment and Infantry (1945-48). The topic assigned for his doctoral thesis was 'Geochemical dispersion related to copper mineralization in Northern Rhodesia'. This work was sponsored by Rhodesian Selection Trust Ltd. This came about because [Alfred] Chester Beatty Jr , at that time Chairman of Selection Trust (he became its President in 1978), was a great supporter of Webb (according to Tooms, pers. comm. 2007, this may have come about as a result of Beatty's interest in the effect of trace elements on his Channel Island cattle). It 'laid the foundation for the subsequent highly successful applications of geochemical prospecting in Zambia' (letter from J.S. Webb to A.C. Beatty, 1969).
John Tooms recalls that when Applied Geochemistry began at the RSM, on a budget of £100, the students were housed in one third of a small room using laboratory balances and other equipment borrowed from the RSM Mineralogy section and, so far as techniques went: 'In '52 the Scandinavian work was never mentioned and our interest was concentrated on the Soviet Union publications which gave much emphasis to stanniferous prospecting. Also on the [University of British Columbia] biogeochemistry publications' (J.S. Tooms, pers. comm. 2007).
John Webb's review of the North American work was presented at a meeting of the IMM in April 1953, by which time arrangements were in hand to hold the first 'introductory course on geochemical prospecting' at the RSM in September that year. The DSIR had agreed to sponsor Lakin, by then Head of the Analytical Research Laboratory of the USGS GPS, to participate in teaching the course. Although not present at the IMM meeting, owing to absence abroad, Sir Frank Dixey (1892FRS 1958), Director of the Directorate of Colonial Geological Surveys, who had previously held senior positions in the Geological Surveys of Sierra Leone, Nyasaland, Northern Rhodesia and Nigeria, was extremely supportive of Webb's ideas and his Deputy, E.S. Willbourn, announced at the meeting that least 14 geologists from the Colonial Geological Surveys would be attending the RSM course (E.S. Willbourn in Webb 1953, p. 463). In October 1953, Webb and Lakin also gave the lecture course in Belgium, France, Germany, Italy, the Netherlands, Norway and Sweden (Webb 1954).
At about the time of Webb's visit to the USA, Bloom at the USGS GPS developed a powerful new field test (known today as the 'Bloom test') for the presence of heavy metals in soils and stream sediments, using an ammonium citrate extractant and dithizone (Bloom 1953(Bloom , 1955. Following the work of Lovering et al. (1950), Hawkes began experimenting with the use of this test with drainage sediment samples taken around the Denver area, where the GPS laboratory was housed. He rapidly became convinced that drainage sampling, based on colorimetric analysis of the fine-grained <80-mesh (<0.177 mm) sieved fraction of samples of the active stream sediments, 'appeared to be a completely new approach to primary prospecting of virgin areas that might lead to spectacular discoveries' (Hawkes 1976, p. 7) and decided to seize the opportunity and resigned from the USGS. Legally debarred, as an ex-USGS employee, from undertaking commercial work in the United States, he could do so elsewhere.
Hawkes decided to embark on a speculative stream sediment survey of the Gaspé Peninsula and New Brunswick, because the Appalachian Belt of Canada contained known mineral deposits and, geologically, it had considerable potential. He was joined in this venture by Bloom, and by Riddell from Canada, who was already familiar with conditions in the Gaspé, having previously conducted water sampling surveys there (Riddell 1952). Almost certainly, because of Chester Beatty's support, Webb secured finance for the project from Selco, a Canadian subsidiary of Selection Trust in London (who had first option on any discoveries; J.S. Tooms, pers. comm. 2007). Fieldwork was only possible during the summer months and in the summers of , Hawkes et al. (1960Fig . 7), with the help of student samplers, completed reconnaissance prospecting over open (non mineral-concession) territory for base metal deposits, with limited follow-up, taking samples from 4937 sites scattered over 27 000 square miles (69 940 km 2 ) of eastern Canada (Hawkes et al. 1960). Using Bloom's analytical method, they located several promising geochemical anomalies which revealed mineralization, although no orebodies were found in the subsequent staking rush. Furthermore, two metallogenic provinces of 'high relief, featured by scattered localities where stream sediments contain over 40 ppm and as much as 200 ppm exchangeable heavy metal' (Hawkes et al. 1960, p. 618), reflecting geochemical heterogeneity, were recognized (Fig. 8); one of which is today known to contain practically all of the significant base-metal deposits. Despite the lack of finding a significant orebody, their survey established the effectiveness of the technique as a low-cost tool for regional mineral exploration, complementing the use of soil surveys as a follow-up method.
However, John Tooms (pers. comm. 2007) says that the term 'regional survey' was never used for that kind of work.

THE GEOCHEMICAL PROSPECTING RESEARCH CENTRE Establishment
While Webb was in Canada on fieldwork in September 1954, Williams was able to announce to the IMM that, following the recognition that there was a need to establish in Britain 'a central research unit similar to the Geochemical Prospecting Section of the US Geological Survey. A Committee, including representatives of the Colonial Geological Surveys, the Geological Survey of Great Britain, the Department of Scientific and Industrial Research, and the mining industry, has recently expressed the opinion that, in view of its pioneering activities in this country, the Mining Geology Department of the Imperial College should act as a nucleus for the prosecution of further research in geochemical prospecting methods' (Williams 1954, p. 51). It is interesting that, in the discussion of his presentation, the economic geologist, Dr. Thomas Robertson (1892Robertson ( -1981, Assistant Director of the British Geological Survey (1949-55) expressed the view that 'a full measure of participation in geochemical prospecting was outside the scope of activities of the Survey, as those were at present defined' but that, as the Survey held a considerable amount of information on the geology of the United Kingdom, he 'hoped geochemical investigators would not be slow to take advantage of [it]' (Robertson in Williams 1954, p. 57). With the help of Sir Frank Dixey and Lord Solly Zuckerman, Baron of Burnham Thorpe, (1904-93;FRS 1943), then Chairman of DSIR, who were both keen supporters of Webb's ideas, the Colonial Office and DSIR provided the finance to establish the Geochemical Prospecting Research Centre (GPRC) in 1954.  As a result of the post-war expansion in numbers of postgraduate students and staff, there was great pressure on space within the RSM, and plans were approved in 1953 to expand it, with the aid of a £330 000 grant from the University Grants Committee, by constructing an additional storey on the north and west wings of the building. This new, Fourth Floor of the RSM building, would accommodate Applied Geophysics (which had been housed in the Physics Department since its inception in 1931) on the north wing and new laboratories for the 'Sub-departments of Applied Geochemistry [the GPRC] and Pure Geochemistry' (the latter under Dr. John R. Butler, appointed Lecturer in Pure Geochemistry in 1954) in the west wing. The new accommodation was ready for occupation in the summer of 1955, and the GPRC moved was established in its new quarters the same year (J.S. Webb in James 1965, p. 1). This was celebrated by an 'At Home ' event, with practical demonstrations, in May 1956(Williams 1963. The goals set out for the GPRC, within the Mining Geology Department of the RSM (Williams 1960, p. 700), as formally established in 1954, were exactly those which had been laid out by Webb in his 1953 presentation (apart from the addition of the text in square parentheses in item (c), above). Webb was appointed as the GPRC's Research Director (Fig. 9) and later became Reader in Applied Geochemistry (1955) and Professor of Applied Geochemistry (1961).
It was initially envisaged that the purpose of the GPRC's 'Analytical Laboratory' would be 'to provide essential services to the field section . . . it is not intended to provide a service to Geological Surveys and mining companies in their normal geochemical work, but only for research purposes in collaboration with the field section of [the GPRC]' (emphasis as in original; Williams 1954, p. 52) and that research into analytical methods would come via the Inorganic Chemistry Department of ICST, under the leadership of the crystallographer and physical chemist, Archibald John Edmund Welch  and the Chemical Research Laboratory of DSIR [National Chemistry Laboratory]. However, in the event, although some advice did come from these quarters in the earliest days (analytical methods for tin and beryllium), chemist Margaret ('Peggy') A. Gilbert (1928-96), was taken on as Webb's personal assistant (letter from J.S. Webb to G.J.S. Govett, March 1959) in 1953 and stayed on as an analyst until 1957. Thereafter, the GPRC academic staff gradually began to enlarge. Webb's principal assistant was John Tooms, who was appointed a DSIR Senior Research Fellow (1955-58) on completion of his PhD; he became an ICST Research Fellow (1958-65) and Reader in Applied Geochemistry .
The whole venture might have ended, unhappily, in the mid-1950s. John Tooms recalled that Peggy Gilbert 'made coffee for Ron Holman, Johnny, me and herself (in beakers), added sugar from the reagent bottle in which it was kept.
However, Ron Holman was very sensitive to cyanide (having been sick from absorbing it through his skin whilst working for a toothpaste company) and stopped us from drinking it. After that, we ceased storing the sugar in old reagent bottles and had bottles of antidotes on the shelf (Potassium cyanide looks like sugar and we were analysing for lead.)' (pers. comm. to G.J.S. Govett, 2008; who remarked that 'it was still stored in the same cupboard as the coffee, etc.' when he was a student a year or so later; pers. comm. 2008). However, Ron Holman could not recall the incident 'even though I do remember Peggy could be quite spiritedly abrasive when sharply reminding all males who was boss in her lab, I don't recall her administering cyanide. Her strictly professional conduct would surely have prevented it! I think it's a good story told down at the pub in South Ken ' (pers. comm. 2008).
In those early days: 'The research was done by post-graduate students, almost all of whom had many years of experience and would bring a knowledge of industry, bush-craft and urgency to each of the one or two new projects in a year. There was no room for failure and it was clear that you had joined the Group; the prime objective was the research, with little expectation of a higher degree. To discover the reality was tough on those who had done well in industry and given up a good job. But if the focus needed to be sharpened, or if there was some doubt concerning the reality, it was to become crystal clear. Not easy, but in the tropical rain forest where the analytical methods were being tested on site (by which is meant on a small wooden box in some tiny clearing in the jungle), and your results were critical, that was one more incentive to make good' (K. Viewing, pers. comm. to I. Nichol 1999).
Details of all publications, theses, and the 'Technical Communications' series (almost entirely written by Stanton and his colleagues, who developed, or adapted, a large number of analytical methods for use in the field; they gave abbreviated operating instructions for those used in the GPRC) are available in the Supplementary Publication. Webb's inaugural lecture, as an Imperial College Professor, 'Applied Geochemistry in Mineral Exploration', was given in June 1962.
Until the publication of Hawkes & Webb's (1962) nowclassic textbook 'Geochemistry in Mineral Exploration', research students were expected to read the work of Jenny  method tested in the laboratory would work in the field, whether rainforest or the outback, for each project was designed to discover the extent of the geochemical dispersion from a mineral deposit, or small mine. The student mapped the deposit and the potential extensions, and collected and prepared the samples. A rudimentary laboratory was arranged under a tree or canvas, the balance and primus stove were levelled, and the reagents and standards prepared on a rudimentary table. The target for the first eight years until atomic absorption spectrophotometry was re-invented and practical by 1963, was to analyse 100 samples per man/day. You had to get that right, you and your dozen or so porters who were away for six months at a time. You were on your own and [had] no need for any additional motivation. ' (pers. comm. 2007).
Once the GPRC became well-established, incoming research students also participated in a field-course in Cornwall, intended for students taking the MSc course in Mineral Exploration (c. 1958-83) or the BSc in Mining Geology. On occasion, it could prove something of a shock: 'My memories are quite vivid because I partook in the winter of 1962/63, which was one of the coldest winters in living memory and it was my first experience of snow, given my life to that point in Western Australia. There was a heavy fall of snow in London a day or two after Christmas and it didn't thaw until March. We took off in late December in a mini-van driven by Ian Nichol and I recall food and comfort stops on the journey when I could barely keep my feet on the packed snow and wondered how I would manage field work. We stayed at a boarding house in Redruth, Cornwall, and set up a lab in the Camborne School of Mines. I recall that Alison McDonald managed the laboratory in Ron Stanton's absence. There were four two-man teams and each was assigned to prospect a segment of the Carnmenellis granite. There wasn't as much snow in Cornwall (to begin with) but I found it physically painful to collect stream sediments from the icy streams. After analysis of the initial batch of reconnaissance samples (tin and copper) in the lab, we conducted cold-extractable copper analysis in the field to trace the best anomaly to its source. Once we reached the cut-off in the stream we set off up a steep slope and found ourselves at risk of serious injury on a jumble of boulders of mullock  [Australian: rubbish; rock from which ore has been extracted], cunningly disguised by a layer of snow. To that point we hadn't detected any evidence of mining in our area, and when we reached the top of the slope, there was an old brick smelter chimney and we realized we had successfully re-located a known lode. I seem to recall that we rounded out the exercise with a soil traverse.
I was teamed with one Peter Binks, an MSc student from northern England, whose later career was spent with Mt Isa Mines, at least partly in WA. At the end of the second or third day in the field we got caught in a blizzard. We worked our way towards the pick-up point, crouching in the lee of a series of stone walls for protection from the driving, horizontal snow and sleet. The end result of this was that we lost our bearings and failed to turn up at the appointed rendezvous. After waiting for a while in the mistaken belief that we were right and Ian had become lost, we caught a bus to some obscure village and holed up in a very cozy pub. We had no means of contacting Ian and expected to have to spend the night there, but Ian found us after an hour or so and took us back to Redruth, somewhat to our disappointment. In short, our performance probably didn't inspire much confidence in our abilities, but the success of the exercise was a very convincing demonstration to me of the potency of geochemical exploration methods' (Richard Mazzucchelli, pers. comm. 2009).
Nevertheless, research in the GPRC was firmly focussed on issues of mineral exploration in tropical areas, particularly in Keith Viewing recalls 'Colonial Africa of the times was untouched, and there was the opportunity. The philosophy was developed too. Deposits of precious and base metals are often within a mineral province and a geochemical province might be found to skirt both, and include for example deposits blind at the surface. The advantage therefore, was inexpensive lowdensity drainage sampling and multi-element chemical analysis to discriminate between the geochemistry of the bedrock and discrete metal deposits. Arc-emission spectroscopy was quick, inexpensive, and semi-quantitative so that much depended upon the interpretation of the results. Small and large scale tests in Zambia revealed traps for the unwary. Significant differences were found between the total and part-extractable metal contents of weathered rocks, and a huge bias could result from the matrix effects of silica and iron in spectroscopy. But successful and large scale tests covered huge tracts of Central Sierra Leone where the rain forest obscures most of the ancient rocks that have a potential for gold, chromium, copper and nickel '(pers. comm. 2007).
The Group 'established the controls to the dispersion of several metals in soils and stream sediments; the products of erosion. First copper in Zambia, then arsenic, chromium and beryllium in Zimbabwe, then arsenic in Sierra Leone. More on copper and cobalt followed in Zambia, and arsenic and molybdenum in the surface waters of Sierra Leone. Research on lead in Uganda was followed by vanadium in Zambia. As one thesis followed another, the demonstration of success was confirmed and the first book on Applied Geochemistry in the English language by Herbert Hawkes and John S. Webb [Geochemistry in Mineral Exploration] was published in 1962. But most of the work was led by John Webb' (K. Viewing, pers. comm. to I. Nichol 1999).
In  , and Webb in Australia, Con Zinc -Rio Tinto Pty. Ltd. of Australia provided two PhD studentships, one in geochemistry (Nicolls), and one in biogeography/botany (Provan). Provan's thesis was supervized by Cole and Webb and his analytical work was undertaken in the AGRG) and Richard Harold Mazzucchelli (PhD 1965); and the Pacific: Eliahu Zohar (MSc 1965). There was virtually no work in the British Isles, other than that of Millman (PhD 1953), C.H. James (PhD 1957) and Peter Richard Donovan (PhD 1965). Successful early results were obtained with secondary dispersion in soil of arsenic, antimony, copper, cobalt, lead, molybdenum, niobium, tin, tungsten and zinc; and copper in drainage sediments (Webb 1958b). Use of these exploration methods soon began to feed through into general mining company practice (Cornwall 1961;Garlick 1961).
Hawkes & Webb's (1962) now classic textbook was subsequently translated into Russian (1964), Serbo-Croatian (1965) and Chinese (1974); a second edition (with Arthur W. Rose, Professor of Geochemistry at Pennsylvania State University, as lead author) was published in 1979. The paper by Govett (2010) paints another picture of what field work in Africa in these early days was like, and that by Butt & Mazzucchelli (2010) discusses the legacy of the direct and indirect influences which Walker, Nicolls, Provan, Mazzucchelli, and later emigrées from the AGRG/GPRC to Australia have had on mineral exploration practice there.

Regional geochemical mapping
When Webb saw the maps of the entire results of the New Brunswick-Gaspé area survey in 1956, he noticed that variations in the background values of the base metal contents of the stream sediments reflected the geology, and he realized that multi-element drainage surveys might eventually be used to produce cost-effective geochemical maps of districts or regions 'showing the distribution of minor elements in relation, not only to mineralization, but to the bedrock geology as a whole. In this fundamental role, geochemical maps would be analogous and strictly complementary to regional geological maps, while at the same time continuing to serve in their more usual technological capacity as a means of mineral reconnaissance' (Webb et al. 1964a, p. 1;J.S. Webb, pers. comm. 1998). This was supported by the results of Tooms' (1955) PhD thesis, and it was decided in 1957 that the GPRC should embark on a series of geochemical surveys in Northern Rhodesia and elsewhere.
In September 1958, at a symposium on the 'Future of Non-ferrous Mining in Great Britain', Webb explained that: 'The small worker has been 'sold' geochemical prospecting, particularly in North America, he believed with some disservice to geochemical prospecting. Tests were very easy and little equipment was required to take samples or make analyses. Unfortunately the interpretation of the data should be done geologically, and in the absence of adequate knowledge such interpretations were often failures. The companies should be expected to do that geochemical exploration that was required on their properties -to detect the bodies for which they were specifically looking. On the other hand, with the large-scale regional service, the development of maps, the distribution of geochemical distribution patterns of all elements over favourable districts as a whole -all that involved considerable effort which was not directly related to finding an ore body and was strictly analogous to the commitment of a [Government] geological survey which accepts responsibility for providing regional maps. It was only a matter of time before surveys must also accept responsibility for providing regional [geochemical] maps ' (updated [September 1958] transcript of talk, Paper 19, pp. 19/23-19/24;AGRG files;Webb 1959, p. 473).
The doctoral theses by Govett (1958), Jay (1959) and Watts (1960 in Northern Rhodesia were accompanied by Kerbyson's (1960) development of suitable spectrographic techniques (see also Govett (2010)). In 1959, Chartered Exploration Ltd. provided the GPRC with drainage samples from three areas of Northern Rhodesia in which they had been working, and these were analysed for a wide range of elements. Taken together with results from elsewhere in Africa and SE Asia, it confirmed the original hypothesis regarding the potential of multi-element drainage surveys.
'A number of aspects of drainage reconnaissance in dissected areas have been, or are being, studied in Northern Rhodesia (Kaluwe, niobium), Uganda (copper-cobalt), Borneo (copper, chromium), Nigeria and Malaysia (tin and associated metals), Burma (lead-zinc) and Sierra Leone (molybdenum, and arsenic associated with gold). In the majority of these areas we have worked mainly on the silt fraction, but in the Malayan, Nigerian and Northern Rhodesian, Borneo (chromium) and Uganda projects we are also concerned with the heavy mineral and other fractions of stream sediments. . . . Since 1958 we have been engaged on a programme of research in collaboration with the Northern Rhodesian Geological Survey, the purpose of which is to examine the applicability of comprehensive regional reconnaissance (geochemical mapping) and the possibility of detecting mineral provinces in this way; the work at Kaluwe was, in fact, part of this programme. Spectroscopic analysis of drainage samples is being used as one of the analytical tools. Regional geochemical mapping is also being investigated in Malaya, Nigeria and Uganda' (letter from J.S. Webb to J.L. Walker, Charter Consolidated Ltd., December 1959).
The Chartered Exploration spectrographic laboratory was subsequently visited by Miss Daphne Ann Poynter of the GPRC analytical staff (Fig. 10), and on her return, she recommended modifications to Kerbyson's technique along the lines of the high-productivity procedures used by Chartered Exploration. Consequently, in 1960 the DSIR provided the money for the additional equipment, purchase of apparatus for the wet-chemical determination of elements likely to be present in concentrations below the threshold for spectrographic determination, and for additional staff.
Keith Viewing comments that Webb's recognition that a geochemical province, analogous to a mineral province, in which some elements would be found to be enhanced within coherent regions that contained as their core, discrete mineral deposits, would be reflected by a regional geochemical map, was a concept of fundamental importance 'first tested in Zambia in reliable conditions and was demonstrated in Sierra Leone in difficult conditions. . . . The patterns were important for the sample density was low, due to cost, and the emission spectrograph in the semi-quantitative mode had a low level of reproducibility. You had found the pattern; go and find the source!' (Viewing pers. comm. to I. Nichol 1999).
The concept was first initiated in 1960 (letter from J.S. Webb to D. Williams, July 1960), using drainage samples collected during a reconnaissance survey for base metals by Namwala Concessions Ltd. over 3000 square miles (7770 km 2 ) of the metamorphic and intrusive rocks of the Namwala-Livingstone area, Zambia, together with maps and additional data collected during their survey, which were made available to the GPRC. The area made an ideal testing ground, as it contained a variety of rock types (Katanga metasediments, gneisses, schists and acid and basic intrusives), a number of mineral occurrences, and considerable variation in both drainage and topography (J.S. Tooms, unpublished note, September 1962; AGRG files). At that time, the only published district-or regional-scale maps were for Nova Scotia, by former GPRC student, Holman (1959), showing copper, lead and zinc in drainage sediments; and by Bell & Overstreet (1960) for copper, lead, zinc and molybdenum in part of North Carolina, but these were intended solely to aid mineral exploration. John Tooms (pers. comm. 2007) comments that the prospecting in Nova Scotia was 'possibly more similar to a regional survey in that a large area was written-off by one geochemist in a couple of months. The area was later the subject of a staking rush which failed to find orebodies.' An initial study by Geoffrey Harden (PhD 1962), of the correlation between the trace element concentrations and geology, including follow-up fieldwork, made it apparent that analysis of the <80-mesh size fraction of the stream sediments would give adequate information. The final maps were based on a subset of the entire suite of samples which, taken from tributary drainage at a density of about c. 1 sample per square mile (1.6 km 2 ), were analysed in 1961-1962, by optical spectrography for 15 elements, plus zinc by colorimetry and cold-extractable (cx) copper and zinc. The stability of these final multi-element patterns was confirmed by a second follow-up survey, in selected areas, undertaken by Nichol in 1963. The results for 10 elements (cobalt, chromium, copper, cx-copper, lead, manganese, nickel, titanium, vanadium and zinc) were published as maps by the Geological Survey of Zambia (Webb et al. 1964a; see also Howarth & Garrett, 2010, fig. 2), accompanied by a separate interpretational report (Webb et al. 1964b).
Not only did the maps delineate favourable areas for further prospecting, but they also yielded information relating to fundamental regional geology and, very tentatively, to human health: Discussions with government doctors suggested that there might be a correlation between areas with higher copper values in the drainage and a reduction in incidences of bilharziasis in the local human population. It was thought that higher copper concentrations might disrupt their reproductive cycle, or kill, the snails which acted as the vector for the disease, but 'no one could be found who had any idea how much copper killed snails, let alone how much was necessary to prevent them reproducing . . . the data were too imprecise to be publishable ' (J.S. Tooms, pers. comm. 2007;Webb 1964, p. 507). Nevertheless, this was a factor in the subsequent broadening of the scope of work undertaken by Webb's group over the years. The effectiveness of the multi-element regional approach, based on stream sediment sampling, was confirmed in subsequent studies, supervised by Nichol, in Sierra Leone (James PhD 1965;Garrett PhD 1966). See also discussion of this early work in Garrett et al. (2009, pp. 206-207).

Agricultural applications
As mentioned in the Introduction, Webb had an early interest in medicine and, in later years, he would become Honorary Secretary (1975Secretary ( to c. 1982, of the Rowhook Medical Society (a campaigning-group founded by the pathologist and virologist, George Dick). In early 1963, with the co-operation of the Irish Geological Survey and the Irish Agricultural Institute, Webb began to investigate the relationship between regional geochemistry and agricultural problems.
A preliminary study in counties Wicklow and Carlow, Eire, found an inverse spatial correlation (Webb 1964, fig. 1) between the occurrence and severity of 'pine' (a disorder caused by a deficiency of cobalt in the diet) in sheep and cattle on soils derived from granite (Walsh et al. 1955), and the cobalt content in 750 samples of the <80-mesh fraction of the stream sediments analysed by Warren John Atkinson (PhD 1967). In September 1963, under a Special Research Grant from DSIR, Atkinson undertook a stream sediment survey in county Limerick, in which seleniferous and molybdeniferous soils were known to give rise to toxicity in grazing cattle. Patterns of elevated selenium and molybdenum concentrations in the drainage sediments were traced to the outcrop of the Clare Shales, a marine black-s hale horizon (Webb & Atkinson 1965;Thornton et al. 1966, fig. 1). Subsequently, a pattern of elevated molybdenum concentrations over the Culm Measures, shales of a similar age in Devon, England, was found to be related to the occurrence of 'scouring' in calves (Webb 1964, fig. 3).
This work raised considerable interest among agricultural scientists, who were keen to see its scope broadened to national levels (letters from T. Walsh, Director of the [Irish] Agricultural Institute and W.G. Alexander, Agricultural Research Council, London, to J.S. Webb, November and December 1964). In order to expedite this, agricultural scientist, Iain Thornton (b. 1934) joined the AGRG staff in 1964 as a Research Fellow. He had taken a BSc in Agricultural Chemistry (1956) and an MSc in Agriculture (1958) at the University of Durham and had then worked as a soil scientist in the Gambia, West Africa . In 1968, he would be awarded his PhD from London University with a thesis on 'The application of regional geochemical reconnaissance to agricultural problems.' The following year, Nichol and Thornton oversaw three regional drainage reconnaissance surveys, encompassing c. 1000 square miles (1610 km 2 ) each, in North Devon, Derbyshire and North Wales. This work was undertaken as part of a Special Research Contract for the Natural Environment Research Council (NERC) through the British Geological Survey, with an additional grant to support the biogeochemical work. Detailed interpretation of the results formed the basis of doctoral theses by William Kenneth Fletcher (agricultural applications, supervised by Webb & Thornton; 1968), Richard Francis Horsnail (mineral potential and geological applications, supervised by Webb & Nichol;1968) and by Jeffrey Khaleelee (computerbased production of geochemical maps, supervised by Nichol; 1969). The resultant geochemical atlases for arsenic, chromium, cobalt, copper, ferric oxide, lead, manganese, molybdenum, nickel, vanadium, tin, titanium, and zinc (Nichol et al. 1970alead, manganese, molybdenum, nickel, vanadium, tin, titanium, and zinc (Nichol et al. , 1970blead, manganese, molybdenum, nickel, vanadium, tin, titanium, and zinc (Nichol et al. , 1971see Howarth & Garrett (2010), fig. 7) were the first of their kind to be produced in Britain.

Marine investigations
Application of geochemistry to marine mineral exploration began in 1961 with Martyn Durant Baker's study (supervised by Webb) of the 'Secondary geochemical dispersion of copper and zinc in coastal environments, Vana Levu [island], Fiji' (no thesis was submitted), in which the distribution of tin and copper from a pyritic ore body on the Udu Peninsula was examined in rocks, soils, stream sediments, estuarine deposits and offshore shallow marine deposits (Fig. 12). Some studies were also made of metal concentrations in waters, vegetation and marine organisms (Webb et al. 1963). In 1963, Pablito Moslares Ong (PhD 1966) began a study of the dispersion of tin in the bedrock, stream sediments, beach sands and unconsolidated marine sediments of Mounts Bay, Cornwall and, because of increasing interest in the possibility of obtaining mineral resources from the ocean bed, in 1964, David Spencer Cronan (PhD 1967) began a study of the composition of manganese nodules recovered from the floor of the Indian Ocean (a topic in which he continued to specialize; see Cronan 2010). Both these projects were supervised by Tooms.

THE APPLIED GEOCHEMISTRY RESEARCH GROUP
As discussed above, by the 1960s, the topics of both GPRC theses (Fig. 13) and subsequent publications (Fig. 14) had begun to broaden considerably in scope. In Webb's Introduction to 'Research in Applied Geochemistry. Applied Geochemistry Research Group Progress Report no. 1' (James 1965 (ed.)), he states: 'During the last three years, in addition to continuing with our work in geochemical prospecting, the research has been extended to other aspects of Applied Geochemistry including regional geochemical mapping, biogeochemistry in relation to agriculture and epidemiology and marine technology. In order to be consistent with the greater breadth of our field of interest, the title of our section was recently changed to the Applied Geochemistry Research Group [AGRG] (my emphasis). This led the author to erroneously state (Howarth 2007(Howarth , 2008) that the AGRG was formed in 1965. However, in research for this paper, evidence from GPRC/AGRG thesis acknowledgements shows that in those submitted before 1965, all acknowledged the GPRC; in 1965, 4/6 used GPRC; in 1966 2/5 used GPRC; and after 1966, they all acknowledged the AGRG. However, the recollections of those former AGRG staff and students whom the author has been able to contact to ask what it was called when they  joined it suggests that the change of name most probably took place in early 1963. During the AGRG years, Webb was a member of the Home Office Forensic Science Committee, 1969-75;Vice-president, 1971-73, andPresident, 1973-74, of the IMM; Regional Vicepresident (Europe) of the Society of Economic Geologists, 1979-81; member of the Board of Council of Engineering Institutions, 1973-74; and a member of the Royal Society Working Party on Environmental Geochemistry and Health, 1979-83.
There were two maxima in numbers of staff and research students in the years covered by this account: 1969-73 and 1979-83 (Fig. 11). These peaks broadly correspond to work on (and subsequent follow-up studies to) the 'Provisional Geochemical Atlas of Northern Ireland' (Webb et al. 1973) and its successor, 'The Wolfson Geochemical Atlas of England and Wales ' (Webb et al. 1978). Figures 13 and 14 show the overall research and publication activities in all the above fields as a function of time (for details of the publications and theses, see the bibliographies which follow). Figures 14 and 15 give a breakdown of publication activity in Environmental Geochemistry and Technical Services. Accounts of work undertaken in the above fields follow in the papers by Cronan, Hale, Howarth & Garrett, Thompson and Thornton (2010).

Geochemical atlas production
The GPRC had carried out geochemical reconnaissance of a number of areas in Zambia, Sierra Leone, Eire and the UK, which totalled over 25 000 square miles (64 000 km 2 ), and Webb felt that the results 'have fully confirmed the basic concept and several important correlations between the geochemical patterns, the geology (including mineralization), soil type and the incidence of agricultural disorders have been disclosed' (Webb 1966, p. 1). By November 1965, AGRG had established that regional geochemical surveys could aid recognition of areas of potential disease in cattle and sheep at both clinical and sub-clinical levels. Webb and Thornton now felt able to recommend to the Agricultural Research Council that although 'much work remains to be done before the full scope of geochemical drainage reconnaissance can be realized. Nevertheless, having regard to present results and the low cost factor, geochemical reconnaissance surveys on the regional scale can now be justified. An additional incentive is provided by the fact that the data so obtained constitute a reconnaissance aid in mineral appraisal, and are also of fundamental geological significance. A further possibility is the application of the results in other fields, including epidemiology and pollution studies' (Webb & Thornton 1965, p. 7).
In the following year, AGRG began a long-term programme to study: the relationship between the occurrence of both nutritionally-essential and potentially-toxic elements and the viability of agricultural crops and livestock; epidemiology; and the identification of the severity of contamination in the urban, agricultural and marine environments as a result of both air-borne and water-borne pollution related to human activity (e.g. Webb 1969Webb , 1970Webb , 1971Webb , 1973Thornton 1973Thornton , 1974aThornton , b, 1977Thornton , 1978aThornton & Webb 1967, b, 1974, 1975aWebb et al. 1966Webb et al. , 1968. This work, which continued long after Webb's retirement in 1979, is discussed by (Thornton (2010).
Initially based on the results of the geochemical mapping of parts of Devonshire, Denbighshire and Derbyshire (Nichol et al. 1970a, b;1971), it was decided to confirm Webb's vision that regional geochemistry could provide an invaluable multipurpose adjunct to conventional geological mapping by undertaking a pioneering stream sediment geochemical atlas of Ireland (Webb 1966). Initial support was forthcoming from the Agricultural Institute, Department of Industry and Commerce, and Geological Survey (in the course of other work 11 000 samples had already been collected and it was only necessary to obtain a further 10 000) but, regrettably, the Department of Finance was unwilling to support the project with additional ad hoc monies, and the idea had to be dropped.
Fortunately, better success was had with NERC, which had already agreed to partly finance a similar survey of Northern Ireland. Some 4800 samples were taken from tributary drainage in 1967 and analysed using an ARL 29000B Quantometer, a 40-channel automatic emission spectrometer (installed at ICST in October 1966), for aluminium, barium, calcium, chromium, cobalt, copper, gallium, iron, lead, magnesium, manganese, nickel, potassium, scandium, silicon, strontium, vanadium and zinc; arsenic and molybdenum were determined by rapid chemical procedures. A limited release of the results for copper, lead and zinc over 1600 square miles (4100 km 2 ) of Counties Fermanagh and Tyrone was made to mining companies in March 1969, to the alarm of the Ministry of Commerce, who were concerned that there could be an increase in prospecting activity before the new Mineral Rights Legislation, which was then being worked on, was finalised (letter from H.E. Wilson to J.S. Webb, December 1968). By June, the maps had been purchased by over 20 mining organisations. Full coverage was released in November 1969 and taken up by c. 40 companies.
Because of the 'noise' present in the data as a result of natural sampling variability and analytical error attributable to the rapid 'fit-for-purpose' analytical techniques used, the maps in 'The Provisional Geochemical Atlas of Northern Ireland' (Webb et al. 1973) were produced by local smoothing of the data (Howarth & Lowenstein 1974), a concept that was already proven in regional geochemistry (Garrett & Nichol 1967). Webb insisted on having the word 'provisional' in the title on the grounds that it was a pioneering effort, and others (such as the Geological Survey) would be bound to produce more detailed geochemical maps in the course of time. It was the first regional geochemical atlas of its type and was well received both by geochemists and cartographers: 'On the basis of sheer originality this atlas must be regarded as an important publication. One is tempted to express the hope that it will be the forerunner of a time when geochemical maps are regarded as naturally as geological maps are now' (James 1974, p. 12). The US Geological Survey was so impressed that they purchased 20 copies (letter from J.M. Botbol to R.J. Howarth, March 1974) and the reviewer's prediction came to pass (Garrett et al. 2008).
The British Geological Survey (then the Institute of Geological Sciences) had begun their own reconnaissance geochemical work in 1968, and their first geochemical atlas, for the Shetland Islands, was published in 1978 (Plant & Moore 1979, p. 109).
Having proved the atlas concept worked, AGRG work now began on 'The Wolfson Geochemical Atlas of England and Wales' (Webb et al. 1978), so-called because, in the absence of support from what might have been expected to be the more usual sources of funding for a national-scale project of this kind, its production was made possible with the help of Lord Zuckerman, Chairman of the Trustees of the Wolfson Foundation, with a £625 000 grant in 1969. The atlas was based on the same principles as used before for the Northern Ireland. Some 50 000 stream sediment samples (Webb et al. 1978, p. 19) were collected in the summer of 1969 and analysed using the Quantometer for aluminium, barium, calcium, chromium, cobalt, copper, gallium, iron, lead, lithium, magnesium, manganese, nickel, potassium, scandium, silicon, strontium, tin and vanadium and by atomic absorption spectrophotometric or colorimetric methods for arsenic, cadmium, molybdenum and zinc. Details of the presentation methods developed for both the Northern Ireland and England and Wales atlases are discussed by Howarth & Garrett (2010).
Webb always regarded the geochemical atlases produced by the Group as of an essentially reconnaissance nature 'carried out primarily for applied research purposes concerned with examining the multi-purpose scope of broad-scale regional geochemical mapping and the considerable problems of handling, processing and presenting the large volumes of multielement data involved. . . . Clearly there is much room for improvement before arriving at a definitive national atlas and the more rigorous regional surveys currently being undertaken by the [British Geological Survey] are the next important steps towards this objective' (Webb 1975, p. 14). As a cost-effective demonstration (a total expenditure of c. £580 000 over 5 years) of the technology, the England and Wales atlas succeeded admirably. Detailed study of the data collected for both geochemical atlases gave rise to a large number of research projects in regional and, particularly, in environmental geochemistry (Figs 13, 14, 16, 17).
The original emphasis of the England and Wales atlas had been on the effects of natural, mining and past industrial pollution on farmland (interpretation of the atlas data in terms meaningful to agriculture was latterly supported by the Agricultural Research Council) and sites of obvious industrial and urban pollution had consequently been avoided; likewise although the Group's Fisheries programme (financed by NERC since 1968) included pollution studies in a number of estuaries selected for their economic potential, heavily polluted rivers and estuaries were avoided. A subsequent programme of fill-in stream sediment sampling was undertaken to obtain further data.
By 1981, multi-element geochemical patterns had been assessed in relation to a number of agricultural disorders (e.g. glutathione peroxidase activity in sheep's blood, an indicator of low selenium status; and molybdenum-induced hypocuprosis in cattle); a comparison had been made between trace element levels in stream sediments and soils (important in the absence of regional trace element maps based on soil); and 'heavy metal' (or, rather, metal and semi-metal) contamination of agricultural soils. From the mid-1980s, there was increasing emphasis on pollution within the urban environment itself (Johnson et al. 1982(Johnson et al. , 1985Culbard et al. 1983aCulbard et al. -c, 1986Culbard et al. , 1988Watt et al. 1983;Culbard & Johnson 1984;Thornton & Culbard 1985;Thornton et al., 1985;Davies & Thornton 1987;Davies et al. 1987a-c;Hunt et al. 1988). Students involved in these programmes  (2010) discusses the AGRG's work on the relation between geochemistry and agriculture, wildlife nutrition, and environmental contamination from mining and smelting; and the geochemistry of the urban environment, a major project begun in 1981.

A time of transition
Webb's concepts of rapid 'cost effective' methods for geochemical reconnaissance and what we now call 'fitness-forpurpose,' regarding the ensemble of research objective, field sampling and, particularly, the analytical system, was far ahead of its time and is only today beginning to be widely applied internationally in a number of other fields. As Thompson (2010) puts it: 'there is no point in paying a premium for high accuracy analysis when low accuracy will achieve the same ends.'  With the passage of time, it became evident that Webb's overall approach worked very well in practice: his aim in regional mapping was always to rapidly attract attention to areas which warranted further, more detailed, investigation, rather than to provide immaculate point-source detail in the first place (a role which Webb always regarded as that of national geological surveys). It was a concept which the mining community had become well-used to.
However, it was hard to convince some in the Geology Department that it was a valid approach, and it was not helped by the fact that those in AGRG were on the Fourth Floor of the RSM building. This sense of distance was echoed by Mary Pugh in a draft of her history (1995) of the later years of the ICST Geology Department: 'Somehow, unfortunately, the Fourth Floor always seemed remote from the rest of the department, maybe it stemmed back to the collective conscience that the Fourth Floor was physically an afterthought being built in 1955. . . . Perhaps because the Group concentrated on research, which was funded by outside organisations and had to operate on a more industrial footing made it different. . . . Ironically however, by the late 1980s the rest of the Department was having to follow the lead set by AGRG.' The patronizing view held by some of the work undertaken in Webb's group is reflected by a comment in the recent centennial 'History of Imperial College': 'Geologists, . . . wanted to occupy all the borderlands of their subject. For example, according to the quinquennial statement for 1972-77, the following were just some of the research areas in the geochemistry section alone: prospecting in desert and glacial terrain, interpretation of geochemical anomalies, geochemistry in mineral exploration, remote sensing techniques, compilation of geochemical maps, agricultural applications of the geochemical survey, significance of trace element distribution in relation to animal and plant nutrition and disease, applied marine geochemistry, and geochemical reconnaissance in pollution surveys. An example of the latter work, which could equally well have been carried out in the chemistry department, is Iain Thornton's work on the study of heavy metal and radioactive pollutants in dusts and soils' (Gay 2007, f.n. 73, pp. 552-553;my italics).
However, this last statement is belied by the fact that urban geochemistry (discussed in Thornton (2010)) was later also undertaken by the British Geological Survey as part of the overall expansion of research and development in this field.
Webb and his staff worked extremely hard to obtain research contracts and had great support from successive Heads of the Geology Department: David Williams (1950-63), John Sutton  and Rex Davis (1974-78 While the GPRC featured prominently in David Williams' unpublished 'History of the Department of Geology ' (1963) up until 1957, and Mary Pugh's continuation volume covering the years 1958-1988, there is virtually no mention of the work done by GPRC or AGRG in Gay's 'History of Imperial College' (2007), nor in a recent pamphlet and internet offering celebrating 100 years of the RSM (Anonymous 2007). The support which Webb, and the work of the GPRC/AGRG as a whole, had received from successive Heads of Department evaporated once engineering geologist [Sir] John Knill (1934-2002 took over as Head of Department from Davis (1979)(1980)(1981)(1982)(1983)(1984)(1985)(1986)(1987)(1988). It became evident that the former encouragement for the AGRG's research work, experienced under Knill's predecessors, was at an end. As a result, some AGRG staff felt it was time for a change and chose to leave, or were absorbed by other parts of the Department; others left as the result of cessation of research funding; the net result was a rapid reduction in personnel (Fig. 11). Webb retired in June 1979, to become a Senior Research Fellow, handing over the baton to Iain Thornton as the new head of the AGRG.
Although Webb (Fig. 18) had the opportunity, having the position of Senior Research Fellow until 1988, to continue active involvement with AGRG following his retirement, he had, in effect, completely withdrawn from it, preferring to spend more time on his lifelong hobbies of fishing, shooting, painting and, perhaps, thinking back to the days when he operated a ham radio. He remained at the family home, Stone Cottage, Slinfold, until he was admitted to the East Surrey Hospital a month before his death.

Aftermath
Knill's attitude to those in the AGRG did not help growing problems with the short-termism of research contract funding. The situation only became worse following the unfavourable findings of a Departmental review by the Earth Sciences Review Committee of the University Grants Committee in 1986-7, which concluded that the Geology Department as a whole, hitherto renowned for its work in applied geology, 'did too much contract work [its contract and external research income was then the third highest in the country] and too much private consulting' and decided that the Departmental budget should be cut by 30% (Gay 2007, pp. 665-666). The official closure of the AGRG took place on 29th July 1988. The environmental programme was renamed the Environmental Geochemistry Research Group and moved for administrative reasons into the IC Centre for Environmental Technology (ICCET), although remaining physically in the Geology Department and using the ex-AGRG analytical facilities. Its work thereafter became entirely focussed on environmental geochemistry research, under Iain Thornton, who had just become Professor of Environmental Geochemistry, a joint position in ICCET (where he was appointed Chairman and Head of Department) and the Geology Department. Marine geochemistry was absorbed into the main body of the Geology Department. Research in both these fields would continue to prosper for another 20 years, attracting funding well in excess of £2M from the UK government, research councils and industry.
Webb had pursued his vision for applied geochemistry over the years with a single-minded tenacity, often in the face of considerable scepticism. 'I am very conscious of the difficulties John experienced in getting his research underway; very original topics at the time that individuals had difficulty recognizing the potential of; the early research outside the U.K. and communication difficult ' (I. Nichol, pers. comm. to R.G. Garrett, 1999). His farsighted vision of geochemical atlases as a strategic national requirement eventually became fully realized in the UNESCO Division of Earth Sciences International Geochemical Mapping Project, inaugurated in 1988. By the time of his retirement, in 1979, technology-transfer to the mineral exploration industry of field and laboratory methods developed by AGRG had been wholly successful. Over 100 students had been trained, many of whom went on to leading positions world-wide in the mining industry, geological surveys, or academia.
A celebratory meeting to honour Webb's life-long contribution to the field of Applied Geochemistry was held at Imperial College on 29th April 1983, at which former students, colleagues and collaborators from around the world were present, some as speakers. The collection of papers was published as 'Applied Geochemistry in the 1980s' (Thornton & Howarth (eds) 1986). In the Geology Department of ICST, his name was commemorated on 6th March 1984 by the formal naming of room 4.59B in the Royal School of Mines ( Fig. 19