240 RUSSELL: THE FIRST BERNARD DYER MEMORIAL LECTURE [Vol. 75 The First Bernard Dyer Memorial Lecture BY SIR E. JOHN RUSSELL (Delivered at the Annual General Meeting of the Society, March 17th, 1950) THE leading scientific societies in our country have developed the praiseworthy custom of periodically suspending their normal activities for an evening and reviewing some aspects of the work of one of their distinguished leaders who, though no longer with us, lives on in the notable things he accomplished. You have decided to commemorate in this way the work of Bernard Dyer, one of the founders of your Society, and your Council has done me the honour of asking me to give the first of these lectures. I am proud to do so : I had known him for almost fifty years and had a deep respect for his high analytical attainments, his complete intellectual honesty and, like others who knew him, had great affection for him in his kindly sympathetic human nature and the friendly accessibility and helpfulness he invariably showed to younger colleagues.He was born in London on February 25th, 1856, the son of J. A. Dyer, then the News Editor of the “Daily News”; after some preliminary instruction he went in January, 1871, to the City of London School, then in Milk Street. The school was in advance of most others in its day in that it had long included science in its syllabus. Formal teaching of practical chemistry had begun in 1847* when Thomas Hall, an assistant writing master, added this to his other duties; later, in 1860, a laboratory assistant, Henry Durham, joined the school and from 1868 was promoted to help in the science teaching.Hall retired in 1868 and was succeeded by Isaac Scarf: it was under him and Durham that Dyer received his first teaching in science. Neither of these men had any university qualifications, but Scarf was described as “a fine experimenter” by G. H. J. Adlam, a later senior science master of the school. There was a certain amount of self-teaching of science in those days (my own father indulged in it, much to the sorrow of my mother, who disliked the evening conversion of our small living-room into a very smelly chemical laboratory), and both Scarf and Durham clearly had scientific knowledge and the power to impart it. The head of the school, Dr. Edwin Abbott, one of the great headmasters of the nineteenth century, had already been in charge for six years when Dyer entered; he was primarily a grammarian and a theologian: his “Shakesperian Grammar” is described as a permanent contribution to English philology, and he wrote also many theological works.]But he encouraged science teaching in the school, in spite of the violent clash between Science and Theology that had broken out at Oxford in 1860 and never quite died down. Darwin, Huxley, Lubbock and Tyndall were all in full vigour. Huxley in particular was ca.rrying on a stirring campaign for the wide- spread teaching of science. A Royal Commission had in 1863 recommended that all boys should receive instruction in some branch of natural science during part of their school life, and Dr.Abbott faithfully carried this out; the science syllabus was broad, not to say ambitious: it ranged over chemistry, physics, botany, zoology and geology. I have been unable to discover any details of the teaching or what textbooks were used: Roscoe’s “Lessons in Elementary Chemistry” was then very popular; it first appeared in 1866 and passed through fifteen editions by 1869; other elementary science books covering the school subjects were in the same series. There was, of course, no organic chemistry: indeed, a distinguished scientist of the time had described it as “a more or less circuitous route to the sink.” Whatever a modern educational expert might think of the appliances and the paper qualifications of the teachers, the results were remarkable. A vigorous intellectual life developed in the school.Dr. Lance Kramer, the present senior biology master, to whom I am indebted for much information about the school at this time, tells me that the school was then conquering Oxford and Cambridge by scholarship winning. But much more important, some outstanding scientific leaders were being produced and the school can hardly ever have had a better group of future scientists than in the time that Dyer was there. F. Gowland Hopkins, five years Dyer’s junior, entered about the same time as he did; W. H. Perkin, jun., was there, as his distinguished father had been twenty years before, and also A. G. Perkin. * The centenary of this event was celebrated by a, brilliant function at the school on November 6th, 1948.May, 19501 RUSSELL: THE FIRST BERNARD DYER MEMORIAL LECTURE 241 Dyer, however, remained only two years at the school; he left in December, 1872, and decided on agricultural chemistry as a career. The choice seems remarkable.Dyer was a Londoner born and bred, with no particular agricultural associations so far as I have been able to discover, and chemistry in those days did not offer anything like the scope of later times. A few consultants were doing some very useful work, but for the young man without capital there was little more than teaching or a post as works analyst from which he might hope for promotion into the works itself. Having decided on chemistry, the choice of agricultural chemistry was not as strange as it sounds: London in 1872 was much closer to the country than it is now.Its population was only 3.3 millions, and including Outer London, 3.9 millions: the fields were only at walking distance from any point in the city, and places like Lewisham, Stratford and others were distinct and separate villages. London also had a large animal population. Refrigerator transport of milk had not yet begun; cows were very commonly kept in the yards and stalls of the retail milk-sellers who, in the main, were small individual tradesmen, the big combines not yet having started operations. Horses were the only source of traction; the mews were actual stables. For all these animals large quantities of hay, straw, grain, roots and other fodders were needed, and farmers’ carts came rumbling in from Essex, Kent and Surrey, bringing these commodities and taking back the animal manure.All these must have been familiar sights to young Dyer on his way to and from school. Whatever the reason of his choice, at the age of seventeen he became for three years pupil-assistant to Dr. Augustus Voelcker,* the leading agricultural chemist in the country, who, as consulting chemist to the Royal Agricultural Society, received in his laboratory in Salisbury Square, Fleet Street, samples of very varied kinds for analysis and report: soils, manures, feeding stuffs, and the many other farmers’ requisites, and he was so closely in touch with practical farm problems that he knew the sort of information his clients required. There were as yet no county agricultural institutes or advisory staffs; farmers in need of advice had to go to their own societies: either their county society, if they had one, or a great society such as the Royal, the Bath and West, or the Highland.For a young man ready to take full advantage of his opportunities the training must have been admirable. Dyer was fortunate too in his fellow-workers: they included Voelcker’s two sons, John Augustus and E. W., Alfred Smetham and F. J. Lloyd. Four of them: Dyer, the two Voelckers and Smetham rose to such distinction that they became Presidents of your Society; when I came into agriculture in 1901 all five names were held in great respect by the farming com- munity. Knowing what Dyer was like in later years, one can imagine that this group of able young men must have had great times both a t work and at play; he clearly was happy there, but as to details, history is silent: Mr.Eric Voelcker has kindly looked through his grandfather’s papers but found nothing.? Then in 1877, when he was only twenty-one years of age, he took his courage in his hands, and set up quite independently as analyst at 17, Great Tower Street, where he remained till the place was destroyed by a German bomb in the great air raid in May, 1941. I shall deal only with his agricultural work, leaving others more competent to deal with his work on food and other subjects.$ He had at the time, so far as I can discover, only one official appointment, that of consulting chemist to the Devon County Agricultural Association, but his ability and his attractive personality clearly stood him in good stead, for he succeeded in establishing himself and maintaining cordial relations with his old friends, the Voelckers.He wrote easily, and, indeed, had a flair for publicity; he published six papers on Rural Water Supplies in the Agricultural Economist for 1878, followed by a rapid succession of papers in the same journal on a great variety of agricultural subjects; besides a series of twenty-one papers on Geological Sketches during the years 1881 to 1883. He ranged from crop manuring to cattle condiments; from home-brewing to disposal of rural sewage: all topics of great interest to farmers; and he wrote shrewd commonsense in language they could understand. In 1880 he became Lecturer on Agriculture at the City of London College, * Mrs. Bolton, Bernard Dyer’s daughter, tells me that he already knew Augustus Voelcker’s sons, John Augustus and E.W. The school register does not show that any boy bearing the surname Voelcker ever attended the school. j. F. Gowland Hopkins, who left the school a few years after Dyer, also spent three years in a consulting chemist’s laboratory in the city, but found the life unsatisfying and unpleasant, so abandoned the profession. It might be interesting to speculate what would have happened had he had Dyer’s good fortune. $ Reference to these is made by Mr. George Taylor in the obituary notice he prepared for the Chemical Society (J. Chem. SOC., 1948, 896-8).242 RUSSELL: THE FIRST BERNARCl DYER MEMORIAL LECTURE [Vol. 75 and gave courses there during the next eight: years. Meanwhile, he was building up a reputation in connection with food and agriculture.More and more farmers turned to him for guidance; in due course he was appointed consulting chemist to the County Agricult‘ural Societies of Essex and Leicestershire, as well a:; to many smaller agricultural organisations. He did not confine his activities to the lalioratoiy; following the example of his great teacher, Augustus Voelcker, he made field experiments on actual working farms and published his reports on them in full detail. The first that I can find were made on turnips at Rusper, near Horsham, in 1882 and 1883,* and their purpose was to compare the fertilizing values of raw coprolites with dissolved coprolites, Le., superphosphate, made from them, in each case with and without dung.The reports are on Augustus Voelcker’s lines; and included fairly full analyses of the manures as well as the so-called total analysis of the soil: the amounts of iron, aluminium, calcium, magnesium, potassium, sodium, phosphoric acid and sulphuric acid extracted by strong hydrochloric acid under specified conditions, together with sufficient agricultural detail about previous cropping and manuring to answer questions the farmer would be likely to put. Field experiments rarely come out quite in accordance with the textbook, and this one di’d not: the raw coprolites gave better results than the superphosphate. Dyer at once recognised that this was due to acidity of the soil: an obvious enough explanation now, when soil acidity has long been a prominent subject, but it could easily have been missed then when few chemists thought much about it.Another o‘bservation which did not pass into general knowledge till years afterwards, when Scott Robertson made his well-known phosphatic trials in Northern Ireland, was that farmyard manure, in spite of its low content of phosphoric acid, is a useful source of phosphate to plants and need not always be supplemented with superphosphate. The combination of farmyard manure and superphosphate was no more effective than either alone. Oats were sown on the same plots in the following year so that the residual manurial values could be ascertained. Three years later, in 1886, he made experiments on the manuring of cabbage on the same farm, and showed the great value of nitrate of soda, especially when used in conjunction with phosphate and salt.In the same year h e began his field experiments in Essex with Edward Rosling, who became a life-long friend. Essex was then changing its system of husbandry : the old carn-growing husbandry was no longer profitable and was being displaced by dairying which required large quantities of mangolds, a new crop to many of the farmers. The problems were set by the Committee of the Essex Agricultural Society-was nitrate of soda or sulphate of ammonia the better fertilizer for mangolds, and should the application be at the time of sowing or later as a top-dressing? The experimental scheme was no doubt drawn up by Dyer-the duplication and systematic order of arrangement resembled those of the Sussex experiments-and after discussion, it was finally submitted to and approved by the Special Experiments Committee of the Royal Agricultural Society.As in Sussex, oats were grown in the following year without further manuring to discover the residual values, if any, of the manures applied to the mangolds. In 1887, basic slag was tried: it was then a new fertilizer and was called “basic cinder”: it contained 17 per cent. of P,O, (37 per cent. of phosphate) and its price was forty-eight shillings per ton; its fineness was sixty-five per cent. This set of mangold and oat experiments was continued €or several years, to test an idea then current among farmers that although artificials might give a heavy crop of mangolds, they did so at the expense of the condition of the soil-the next crop, it was thought, would surely suffer.Dyer showed this was not so: the heaviest crops of mangolds were generally followed by the heaviest crops of oats; even heavy dressings of nitrate of soda given to the mangolds benefited the succeeding oat crop, probably, as he surmised, because they produced a large weight of leaves which, being ploughed in, constituted a valuable green manure. These were busy years for Dyer, for besides fulfilling his numerous professional engage- ments and building up his extensive practice, he continued his academic studies, attending courses at the Pharmaceutical Society, King’s College and the Royal College of Science; in 1886-nine years after he had set up on his o’wn account-he sat for and was awarded the BSc. degree of the London University.The records of the Sussex and Essex field experiments ended in 1891. Three years later (1894) he began, with F. W. E. Shrivell, Golden Green, Hadlow, Kent, an extensive series The plots were in duplicate, systematically arranged. Here too the mangolds were followed by unmanured oats. * J . Roy. Agric. SOC., 1884, Vol. 20.May, 19501 RUSSELL: THE FIRST BERNARD DYER MEMORIAL LECTURE 243 of trials on the manuring of market garden crops, aided by a grant from the Chilean Nitrate Committee; these were continued for a number of years and summarised in his book, “The Manuring of Market Garden Crops,” 1904. The feature common to all his field experiments is that they invariably dealt with definite questions of practical importance put to him by the farmers themselves.Dyer devised and carried out simple straightforward experiments that would give the answer. He wrote the report in language which the farmers could understand. As one of them said about his interesting paper on Catch Crops at the Farmers’ Club in November, 1888: “It is wholly practical; it commenced without any introduction and finished without any peroration,” and J. A. Voelcker, in commenting on the discussion that followed, added: “Our scientific men have become practical and I really think that you practical men are becoming scientific.”* His success among the farming community arose from his shrewd interpretation of the results of the experiments and of his own observations and gleanings on actual farms. Very few of his judgments would require much correction in the light of subsequent knowledge. His attitude to grassland in 1886 is entirely modern: “there is perhaps no crop more amenable to good treatment than grass,” he said, “and none more apt to be neglected.”f The times were hard for farmers, and he always gave a financial summary of the results, with shrewd comments: he pointed out, for example, that, if the increased crop only just paid for the fertilizer and the additional cost of handling, there was still a profit because the overheads remained the same, and therefore the cost per ton had fallen.In all these reports Dyer published the analysis of the soil as made by the old method of hydrochloric acid extraction: the so-called “total analysis.” It is significant that he does not mention the analytical data in any of his discussions, and, indeed, looking back over the figures, it is difficult to see what he could have said about them.The P,O, content of his soils varied from about 0.1 to over 0.2 per cent., but the variations had no relation to the responses of the crop to phosphatic fertilizer. Some better method was obviously needed, and as early as 1884-two years after starting his field experiments in Sussex-he began to study the general question of solvents in soil analysis; after a good deal of intermittent work, some of which he states quite candidly was tentative and disappointing, he set about a more systematic enquiry. It had early been recognised that plants can take up only a small fraction of the potassium and the phosphate present in soils.Liebig, in his classical “Chemistry in its Applications to Agriculture and Physiology” published in 1840, had discussed the breaking down of the comparatively insoluble “alkali compounds” of the soil to a more soluble state. The final solution was, he supposed, effected by acetic acid excreted by the plant root and the dissolved material then entered the root. It was, however, Daubeny: of Oxford who first clearly distinguished what he called the “active” potassic and phosphatic plant nutrients from the “dormant” compounds in his classical memoir “On the Rotation of Crops and on the Quantity of Inorganic Matters abstracted from the Soil by Various Plants under Different Circumstances,” and he suggested that the “active” nutrients could be determined by extraction with a solution of carbon dioxide, thus simulating the relations between the plant and the soil.This brilliant suggestion unfortunately could not be carried out, no adequate chemical technique being then available. Unfortunately, also, this memoir did not receive the attention it deserved, and the subject was never properly followed up: Dyer does not seem to have known about it. But the distinction thus emphasised passed into general knowledge, the names adopted being “available” and “unavailable” plant food: this simple division into two groups accorded with the nineteenth century predilection for big, broad generalisations. Liebig’s son, Hermann, had in 1872 examined some of the Rothamsted soils and showed that dilute acetic acid extracted only a little potash from the continuously unmanured Broadbank soil, but twice as much from the soil that had received potassic fertilizer on farmyard manure.P. P. Dbhbrain also used acetic acid in examining the Grignon soils and showed that it distinguished soils responding to phosphatic fertilizer from those that did not.$ Other Continental observers made similar observations. Hellriegel and Wilfarth’s proof in 1886 that this was fixed from the air by the nodule organisms was not yet widely accepted. Dyer had quite an open mind on the subject. * Some of the discussion had turned on the source of nitrogen for leguminous crops. -t “Livestock Journal and Agricultural Gazette Almanac” for 1886. C. G. B. Daubeny, Phil. Trans., 1845, 179-253. Ann. Agron., 1881, 6, 392-3; 1892, 17, 445-54.244 RUSSELL : THE FIRST BERNARD DYER MEMORIAL LECTURE These facts were no doubt well known to agricultural consulting chemists, but were not used by them; it was assumed that the available potassium and phosphate would have some relation to the total, and if this was high in amount it might be assumed that the “available” material would be adequate in quantity.It was Dyer’s great achievement that he broke away from this view and devised a workable method whereby the analyst could discriminate between the available and the non- available mineral nutrients in the soil. Characteristically enough he was not influenced by Justus Liebig’s suggestion that plants obtain their minerals by a supposed secretion of acetic acid from their roots: he may even have overlooked the suggestion, as he does not mention it.Instead, he followed up work that had been started on the Continent on the evaluation of the new phosphatic fertilizers containing no water-soluble phosphate which began coming into use in the early 1880’s; in America, the so-called precipitated phosphate; on the Continent, finely ground mineral phosphlate; and both there and here thephosphatic slag produced by the Bessemer process for making steel. Some selective solvent was needed to discriminate between the more easily soluble phosphate readily available to the plant, and the less readily soluble, and therefore, less valuable phosphate. Ammonium citrate was tried; it seems to have answered well enough for. precipitated phosphate, but was not so good for other materials.Tollens suggested citric acid* as the solvent, and this was strongly supported by Stutzer, who confirmed that in 1 per cent. solution it gave results far more in accordance with field experience than the strong ammonium citrate solution then commonly used, The choice both of acid and of strength was purely pragmatic: the reagent worked well, therefore it was the one to use. It was with this background that Dyer approached the subject of soil analysis. Here also the problem was to distinguish the available fraction of the phosphate from the total. He tried various strengths of ammonium citrate and of citric acid and found no constancy in the amount of P,O, extracted: this varied with the solvent and its concentration. So he started with the assumption then widely accepted by plant physiologists, including Sachs, their great leader, that, as Dyer put it, “plants help themselves to a part of their mineral food by means of the solvent action of their acid root sap on the particles of soil with which the rootlets come into contact.” He collected a hundred plants of twenty different natural orders, separated their fine rootlets, estimated their moisture content, crushed and then boiled them in water, estimated the acidity of the extract in terms of hydrogen by titration with phenolphthalein as indicator, and then calculated the results back on the moisture content, so arriving at the percentage acidity of the root sap.He frankly admitted that the value of the results depended on the efficacy of the extraction process, which he therefore described in considerable detail.The range of values was from 0.003 to 0.014 per cent. in terms of hydrogen, but the value he adopted was 0.013 per cent. Expressed in terms of citric acid this came to 0.91 per cent. Assuming, then, that citric acid was to be used as solvent, the strength should be 0.9 per cent. Dyer rounded this off to 1 per cent., influenced no doubt by Stutzer’s results with phosphatic fertilizers. Partly, he said, because it is “an organic acid, and in that sense kindred to other root sap acids,” partly because “it is the acid generally used by those who have attempted to determine available phosphoric acid in manures,” and partly because “it is at hand in every agricultural laboratory in a state of purity, and therefore a convenient acid.” He claimed no special scientific basis for it: “since the choice of a solvent for use in soil analysis must in the end be empirical, both as regards its form and its strength, it seemed now worth while making some effort to test the expediency of adopting a 1 per cent.citric acid solution.” Fortunately he was persona grata with both Lawes and Gilbert, and was allowed in 1889 to draw samples of the soils from the Hoos field barley plots of which the manurial and agricultural history was well known. The soils that had received no phosphate for thirty-eight years, and on which the crops were suffering from obvious phosphate starvation, contained no less than 2610 lb. of P,O, per acre (0-087 per cent.), while that of the soil manured with superphosphate contained over 4100 lb.(0.14 per cent.): a marked difference certainly, but not capable of explaining the starvation on the unrnanured plots. One per cent. citric acid, however, extracted only [Vol. 75 But why did Dyer choose citric acid? * A. Grupa and B. Tollens, B e y . dlsck. Ckem. Ges., 1880, 13, 1267; v. Ollech and B. Tollens, Jouvn. f Land., 1882, 30, 519.May, 19501 RUSSELL: THE FIRST BERNARD DYER MEMORIAL LECTURE 245 0.005 to 0408 per cent. of P,O, from the plots without phosphatic manuring-160 to 230 lb. per acre-but about six times as much4.04 per cent., lo00 to 1200 1b.-from those that had received phosphatic fertilizer. Similar results were obtained with potash determinations : the “totals” obtained by hydrochloric acid extraction (which are much less than the true totals obtained by fusion methods) were meaningless; the figures obtained by extraction with 1 per cent.citric acid accorded with the agricultural history. Various other deductions were drawn by Dyer from the analyses, perhaps the most important being the confirmation that the major part of the phosphoric acid supplied in the mineral fertilizer remains in the top nine inches of soil, and much of it in a form soluble in 1 per cent. citric acid. He was never able to account for all that had been supplied, and had to assume either that it had disappeared into the sub-soil, or had changed into a less soluble form-this latter is the view now held. He also concluded that the “alkali salts” had rendered some of the soil phosphate more soluble, or, as he later thought more probable, that the manurial phosphate entered into some sort of combination with the alkaline bases and remained in a less insoluble state than if these had been absent; also that the sodium in nitrate of soda had increased the solubility of the soil potash enabling more to be taken up by the crop, but also increasing the loss to the sub-soil.But he was careful to qualify all these deductions by pointing out that they involved the assumption that the different plots had been chemically fairly uniform at the outset. The important practical outcome of this work was his conclusion that a soil containing less than 0.01 per cent. of P,O, soluble in 1 per cent. citric acid solution might be presumed to be in need of phosphatic manuring, while if it contained 0.03 per cent.it presumably was not. If the per- centage of K,O was 0.005 or less, potassic manuring would presumably be necessary for cereals, if 0.01 per cent. or more, it presumably would not. This paper formed the major section of a thesis on which the University of London, in 1892, awarded him the degree of Doctor of Science; it was later (1894) published by the Chemical Society* and attracted a good deal of attention. Lawes and Gilbert realised the value of the work and invited Dyer to make a similarly full examination of the soils of the famous Broadbalk wheat field. He did so, and again the results obtained by the citric acid method accorded well with the field experiments. This paper was presented to the Royal Society, and so much importance was attached to it that it was published in their Philosophical Transacti0ns.t Lawes and Gilbert also invited Dyer to go to the United States in 1900 to deliver one of the periodical courses of lectures for which Lawes had made provision in the Trust Deed of the Kothamsted laboratory.$ These lectures presented the whole of the data accumulated on the Rothamsted soils, and which Lawes had long wished to be brought within “one pair of covers” as he put it.Most of the material had centred round the nitrogen problems, always Gilbert’s chief interest, and here Dyer was little more than the reporters: Gilbert laid down the lines-he was somewhat of a martinet-and all Dyer’s tact was needed to keep things going smoothly. He did so, but he once confided to me that it had been no easy task.Dyer’s own work on the phosphoric acid and potash content of the soils fills some thirty-four out of the hundred and eighty pages; it was, of course, already known in the United States, but the Department of Agriculture was then in the charge of Milton Whitney, who attached little importance to the chemical composition of the soil, but much to its physical properties. In 1890, Mr. Goschen, the then Chancellor of the Exchequer, had put a tax on whisky, the proceeds of which might be used by a county for advancing agricultural education. Several counties took advantage of this and set up agricultural colleges or institutes, and the staffs, anxious to establish contact with farmers, undertook analysis of soils, fertilizers, feeding stuffs, etc., which had hitherto been done by consulting chemists.There were some complaints about this State-aided competition : whether it took business away from the profession or whether For root crops, however, he expected the limits would be higher. Dyer’s Chemical Society paper had appeared at an opportune moment. Trans. Chem. SOC., 1894, 65, 115-67. t Phil. Trans., 1901, 194B, 235-90. $ These had to be discontinued later : at first because material and lecturers were lacking, and afterwards $ Dyer had, of course, special knowledge of this work, having contributed an important paper to the when these abounded, because the financial provision had become wholly inadequate. Chemical Society in 1895 on the Kjeldahl method of determining nitrogen.246 RUSSELL : THE FIRST BERNARD DYER MEMORIAL LECTURE it created an entirely new clientele we need not now discuss; the fact remains that these new advisers wanted analytical methods, and Dyer’s citric acid test appealed very much to them.They were living and working among farmers; their advice was critically received and severely tested, and while at times they received credit for any successes, it was long before they heard the last of any failure. A. 11. Hall and F. J. Plymen at Wye, and T. B. Wood* at Cambridge, all used Dyer’s method and found it worked satisfactorily. Hall and Plymen tried other dilute acids: as no two extracted the same amounts of P,O, and K20 they gave up the old idea of two groups of compounds, one available and the other not available, and regarded the soil as made up of a large number of compounds of varying degrees of solubility and availability, but with no sharp boundary anywhere.Dyer agreed; he did not like the expressions “available” and “una\ailable” plant food in soil; he regarded them as over-simplifications. Turning to the practical question of finding a convenient solvent, Hall and Plymen pointed out, however, that none was so useful as citric acid: others usually put the soils in the right order but extracted 1e:;s P,O, than did citric acid, making the deter- mination more difficult. It was in any case tedious enough; the extraction was in Winchester bottles and lasted a week; the bottles had to be shaken daily, and Dyer had spoken of 4-00 shakes in all; the P20a was estimated gravimetrically by 0.Hehner’s method-but in those days men worked long and cheerfully, and were thankful for a method that worked well. I t was universally adopted at all agricultural institutions. He was now caught up on another line of work. The fertilizer and feeding stuffs trades had not been free from fraud-when I first turned to agricultural chemistry in 1901, one of my revered seniors assured me that, in the lump, all fertilizer people were rogues-and in 1893 the Government passed the first Fertilizers and Feeding Stuffs Act. As advisory chemist to many agricultural bodies, Dyer became intimately acquainted with the working of this Act and the very limited extent to which it met the farmers’ problems. He was a member of the Government Advisory Committee in connection therewith, and was consulted in regard to the amendments and changes that became necessary for the framing of the new Act in 1906, of which he gave a good account in his little book written for farmers, “Fertilizers and Feeding Stuffs”; this passed through a number of editi0ns.t The 1906 Act was not entirely satisfactory, and a new one was enacted in 1926, in the framing of which Dyer played an important part.I am not qualified to discuss these Acts and their bearing on the relations of the farmers and the analyst, but fortunately this is not necessary as it has already been done by your President, Mr. George Taylor, in his very interesting -Streatfeild Memorial lecture before the Royal Institute of Chemistry in 1947. Meanwhile, investigation of the soil itself, quite apart from any immediate practical problem, had been quietly going on in Europe, the United States and this country; it was extended here when in 1906 the Goldsmiths’ Cornpany endowed a research post at Rothamsted, and still more after Mr.Lloyd George set up the Development Commission, which from 1911 onwards made grants for agricultural research, and which later blossomed out into the Agricultural Research Council. The extended investigations thus made possible showed the great complexity of the phosphorus and potassium compounds in the soil. Bassett’s early observation3 that in neutral conditions most of the inorganic phosphorus of the soil is combined with calcium in the form of hydroxyapatite (C~P,O,),Ca(OH), was broadened by D. McConnell,$ who pointed out that the OH could be replaced b:y F, Cl, CO,, 0, or to a small extent by SO,, and the Ca by Mn, Fe and others, the lattice.remaining of the same general form. Other calcium phosphates also exist in the soil as do phosphates of aluminium and iron, especially in acid conditions. Some phosphorus also is held more or less firmly by the soil colloids, and a significant proportion is an organic combination. It has long been known that some changes take place in the soil whereby soluble phosphates become insoluble, for plants are able to extract from the soil about twenty or twenty-five per cent. of the added phosphate; [Vol. 75 Dyer, having safely launched the vessel, left it to follow its course. * A. D. Hall and F. 1. Plymen. Trans. Chem. Soc., 1902, 81, 117-44; A.D. Hall and A. Amos, Ibid.. 1906, 89, 206-22; T. B. Wood; Ibid., 1896, 69, 287; T. B. Wood-and R. A. Berry, J . Agvic. Sci., 1906, 1; 114-21. t The first edition had appeared in 1894 (Crosby Lockwood): it was founded on a series of newspaper articles which had appeared in several journals and widely circulated in different parts of the country. The 6th edition appeared in 1910. Trans. Chem. SOL, 1917, 111, 620. 3 Amer. Man., 1938, 23, 1.May, 19501 RUSSELL: THE FIRST BERNARD DYER MEMORIAL LECTURE 247 what becomes of the rest is not known. The potassium compounds in the soil are, if anything, even more complex in their range. Moreover, the processes by which the plant roots take up their nutrients from the soil are much more complex than was at first thought.The old idea that they excreted from their roots an acid which attacks insoluble phosphates and potassium compounds forming soluble substances which they then absorb was displaced by the view* that only CO, was evolved from the roots. This too has had to go; it is now known that there are other secretions, even it may be acid secretions. Plants vary in their power to absorb phosphates; what is “available” to one may not be “available” to another. Lupins are among the most efficient and Prianishnikov has shown that they help others growing along with them to obtain more phosphate from the soil, apparently by making it more soluble. Further, the plant root is not simply an absorber of dissolved substances: it can also effect exchanges with soil colloids.It would be interesting but would take far too long to set out the present positionof our knowledge of these two interacting complexities, the soil as the source of the plant nutrients and the plant roots as the absorbing agents. Our concern this evening is with the prospect of finding an analytical method that will give some measure of the amount of nutrients that crops can under normal conditions obtain from the soil, remembering, however, that this is not a fixed quantity but varies to some extent with the conditions. Dyer’s method came into wide but not universal use; other solvents in steadily increasing numbers were proposed. No two of these gave the same results: and for any solvent the amount of P20, extracted depended on the concentration of the solvent and the conditions of extraction.This problem was taken up at Rothamsted by J. A. Prescott and myself,? and we found that two reactions proceeded simultaneously when dilute acids acted on soils ; the acid dissolved out phosphate, but the soil slowly absorbed it from the solution. This back action was eliminated by using a diffusion technique, and then it was found that citric, hydrochloric and nitric acids in tenth normal concentration all extracted the same quantity of P20,, whereas by the ordinary analytical process the citric acid extracted nearly twice as much as the hydrochloric and fifty per cent. more than the nitric acid. The citric acid had done this because it had reduced the absorption of the P,O, by the soil. The absorption phenomena fitted the Freund- lich equation then in vogue, and we adopted a physical explanation, though it was afterwards pointed out by E.A. Fisher that a perfectly good classical chemical explanation could be given.$ The analytical data thus recorded the difference between the amount extracted by the acid, and the amount re-absorbed by the soil. Only the direct action is wanted by the analyst; the second upsets his results. So long as he is dealing with similar soils he may assume that the reverse action is also somewhat similar so that his results will still be com- parable, but when he is dealing with different soils the reverse action may be different and he may obtain different analytical results, even though the same amounts of P205 had been extracted. We concluded, therefore, that soil analysis could be helpful when used in con- junction with a soil survey showing the areas of comparable soils, but that standards applied to one soil would not necessarily be applicable to another.The soil survey should be accompanied by field experiments to show how crops on selected soils react towards fertilisers, and these soils being analysed become standards against which the analyst can measure other soils. This procedure is now widely accepted in principle, though not always adopted in practice. Citric acid long retained its popularity because it extracted easily weighable amounts of P205 and of potassium from soils. This advantage was lost when the modern turbidimetric, colori- metric and spectrographic methods came into use. The extremely sensitive molybdenum blue reaction worked out by DenigPsS revolu tionised P,O, determinations by enabling very small quantities to be estimated with sufficient accuracy, while the flame photometer enables equally small quantities of potassium, sodium and calcium to be estimated.Alternatively, potassium can be estimated colorimetrically by the cobalt blue colour developing when Curves expressing the results could be fitted by no ordinary equation. Modern developments in technique have profoundly changed the whole subject. * F. Czapek, Jahr. wiss. Bot., 1896, 29, 321. t J . Agric. Sci., 1916, 8, 65-110. $ Trans. Farad. SOC., 1922, 17, 305-316. 5 Comfit. rend., 1920, 171, 802 (SnCI, test).248 RUSSELL: THE FIRST BERNARD DYER MEMORIAL LECTURE [Vol. 75 ammonium thiocyanate reacts with the cobaltinitrite precipitate dissolved in nitric acid.* It is difficult for the young chemist of to-day, surrounded by mechanical aids and semi- automatic devices, to realise what a profound difference they have made to chemical work.As a student I was brought up on Otto Hehner’s method of determining P206: we always hoped for about 100 or 15Omg. of the ammoinium molybdate precipitate for comfort of weighing: this meant starting with 3 to 6 mg. of P20,. Later, instead of weighing the precipitate, we dissolved it in standard alkali, titrating the excess and arranged our solutions so that 1 ml. was equal to 1 mg. of P206; this was quicker but little more delicate. But in present-day colorimetric methods the quantities measured range between 0.002 and 0.01 mg.(2 to 10 micrograms). In ordinary advisory and routine work 0.005 mg. of P20, and 0.040 mg. of K20 can be determined. The choice of so1ve:nt nowadays, therefore, depends on the con- venience of working, and citric acid, which involves long extraction, evaporation and ignition, is at an obvious disadvantage. Any solvent, even water, can extract enough material for modern methods of measurement. However, like many other scientific advances, this has not led to simplification; on the contrary a bewildering array of solvents is now in use and it is rarely possible to discover why one rather than another is adopted. Mr. G. V. Jacks, Director of the Commonwealth Bureau of Soil Science, has kindly furnished me with a formidably long list, drawn up by Mr.Brind, of methods in use in the different countries. Of the various acids for estimating available P206, 1 per cent. citric acid is still used in this country and in the Dominions, though 2 per cent. is preferred in France, the U.S.S.R. and Java. Acetic acid is much used: 0.5 per cent. in this country and in Florida for the Everglade soils; Spurway’s 0.025 N elsewhere in the United States, also Morgan’s buffered solution of sodium acetate and acetic acid at pH 4.8. Truog’s sulphuric acid 0.002 N and ammonium sulphate buffered to pH 3 is widely used, 0.2 N sulphuric acid has been adopted in Denmark in place of the former nitric acid at a final pH of 2.5; elsewhere hydrochloric acid in various strengths, e.g., 0.75 N (S. F. Thornton of Purdue), 0.3 N in the Middle West of the United States and 0.05 N at Guelph. H.Egnkr’s calcium lactate, 0.028 N , buffered at pH 3.7 with 0.01 N HClt is much used in Germany, Finland, Holland and Russia and New Zealand; but in Germany, Riehm’s modifica- tion, having twice Egner’s concentration of lactate is preferred: it is said to be better for calcareous soils, and it has entirely displaced the citric acid methods formerly used there and also the Neubauer seedling method.$ With the increasing adoption of delicate estimations, water and carbonic acid are being increasingly used: Machigin claimed that carbon dioxide gave the best results with the carbonate soils of Central Asia; it is also used in Holland and in the United States for field tests on calcareous soils. Other analysts have given up acids and use alkalis, especially for calcareous soils: potassium carbonate in India., Russia and South Africa, 0.5 N KOH in Kenya on the fertile red soils.In the case of potassium the problem is more difficult because this element occurs not only as complex silicates, some of which are attacked by dilute acids, but also in combination with the soil colloids from which it can be dislodged by exchange with another base. This exchangeable potash is certainly important in plant nutrition. Two kinds of extraction reagents are adopted: various acids, e.g., 5 per cent. glacial acetic acid at Guelph, or those used for phosphate determinations, Egnbr’s or Riehm’s reagent in Germany; and solutions of neutral salts, e.g., 10 per cent. ammonium chloride to bring out the exchangeable potash.In England several solvents are used. The citric acid method, speeded up by shorter shaking and colorimetric measurements is still widely used, but 0-5 N acetic acid is preferred in the wetter regions of the north and west. Water or Morgan’s acetic acid and sodium acetate solution are also used. I t is doubtful whether the same methods for both phosphate and potash are adopted This does not yet exhaust the list. In no country is there any uniformity of procedure. * More usually at present the potassium is precipitated with sodium cobaltinitrite and estimated by titration with permanganate or ceric sulphate. -f Medd, Centr. Anst. Fiirsiiksv. Jordbr. (Stockholm), 1932, 426. The German methods are described in “Landu.. Versuchs- u.Untersuchungsmethodik (Methoden- buch) .” Die Untersuchung von BGden, I1 Auflage, R. Harrmann, Radebeul und Berlin, 1949, Verlag Neumann. For United States methods see U.S.D.A. Circ. No. 757, 1947, “Methods of Soil Analysis”; also Misc. Pub. No. 259 (chemical quick tests) and “D:iagnostic Techniques for Soils and Crops,” American Potash Institution, Washington, D.C., 1949. Mr. Jacks tells me that the Commonwealth Bureau of Soil Science has in hand the preparation of a monograph describing all these methods. Besides all these chemical methods there is a long list of biological methods which lie outside the scope of this lecture. Bd. 1.May, 19501 RUSSELL: THE FIRST BERNARD DYER MEMORIAL LECTURE 249 in any two of the provinces into which England is divided by the Ministry of Agriculture.Discussions to secure uniformity have led to no results. This is in many ways unfortunate, though less serious than might at first appear. I t is not claimed that the analytical figures have any particular scientific interest : their purpose is to help the adviser, and any procedure giving useful information can be justified. The modern methods of colour, turbidity and flame estimation are not only rapid, but some at least lend themselves to nearly automatic recording. Already direct reading pH outfits are available, mainly based on the glass electrode; last autumn I saw in Vienna the beautiful Zeiss instrument where the analyst had little more to do than put in the soil, and the pH value was then thrown on the screen. The flame photometer used for determining K, Na, Ca, etc., has been developed by Zeiss in Germany and by Beckman in America: the solution is atomised into the flame, the relevant section of the spectrum is selected by suitable filters and the intensity measured by photo-electric colorimeters which can be calibrated in any desired units. Colorimetric and turbimetric determinations are made with an absorption colorimeter, such as Spekker’s absorptiometer (Hilger), using appropriate filters. The next stage seems obviously to be automatic recording and perhaps it is not too fanciful a dream that some of you may live to see direct reading appliances linked up with automatic type- writers, so that the analyst will need only to put the soil into one end of the machine and the typed report will come out at the other.These new methods, along with the automatic balances and automatic pipettes, have speeded up analytical determinations enormously and have vastly increased the number of analyses that can be undertaken. Dr. T. W. Walker, the Advisory Chemist to the West Midland province, tells me that in his laboratory one hundred and fifty soils are analysed daily for pH, lime requirements, available phosphate and potash. In Germany, team processes of the conveyor belt type have been worked out whereby four hundred phosphate determinations per day can be made with the help of three workers, one washer and one trolley pusher, and eight hundred with a second shaking machine and two or three more workers.* Professor G. W. Robinson, lately in charge of the small advisory area of North Wales, reported on twelve thousand soil samples in his last year of office.Mass production of soil analytical data is not attempted in the United States on the same scale as in pre-war Germany, though some of the State Experiment Stations make several thousand soil tests in a year: on the results, Missouri bases some of its recommendations for the manuring of cotton, and Illinois its recommenda- tions for manurial treatment where high yields of maize are desired. Some of the American investigators have devised not only rapid methods, but dwarf apparatus, making a suitcase laboratory for field work, but the big apparatus has prevailed. It has thus become extremely easy to amass vast numbers of figures in soil analysis: but how is one to interpret them? Taking the general run of results the responses of fertilizers are higher in soils indicated by analysis to be deficient in the corresponding nutrient.This is well shown by E. M. Crowther’s investigations on the growth of sugar beet (Table I). TABLE I ANALYSES BY CITRIC ACID METHOD AND FERTILIZER RESPONSES IN SUGAR BEET EXPERIMENTS, 1936 TO 1946 (E. M. CROWTHER) Soil analysis P 0 mg. per 100 g. (soils wit{ less than 5% CaCO,) P 1 8 18-28 28-40 41-1 72 GO, mg. per 1OOg. 3-6 6-8 8-1 1 12-26 Number of experiments 54 54 54 54 63 61 62 62 Additional sugar, cwt. per acre for superphosphate 3.7 1.6 1.1 0.5 for muriate of potash 4.7 4.4 2.0 0.8 But the chemist is commonly called upon to advise on individual soils rather than on large groups of soils, and here the indications of analysis may not always be borne out in practice.H. Rheinwald and G. Constantin, Bodenk. PJ. Emahr., 1939, 16, 1-12.250 RUSSELL: THE FIRST BERNARD DYER MEMORIAL LECTURE Wol. 75 The analyst who is working on mass production lines cannot know the soils as individuals. The problem is made more difficult by the circumstance that the supply of soil nutrients is only one of the factors in soil productiveness, and may not be the chief. American experts have expressed the view that soil analysis may furnish about ten per cent. of the information needed by an agronomist for making an intelligent recommendation regarding a soil manage- ment programme. This ten per cent., however, may be very valuable, even indispensable. Production of valuable crops such as potatoes, sugar beet, market garden crops, involves heavy manuring which in ordinary practice is not to any great extent determined by the soil nutrients, while intensive livestock production yields large quantities of manure enriched with potash and phosphates from imported feeding stuffs, also without regard to the com- position of the farm soil.It is easy to become so absorbed in the elegance of the methods and the beauty of the apparatus that one is apt to forget that the figyres have little intrinsic value: their use is for comparison with known or standard soils, so far as other conditions permit comparison to be made. The greatest value of soil analysis is realised when it is done in conjunction with field experiments, best of all a systematic series such as those in the old German Feld- verszcchsringe, where a large group of fanners carry out uniform field trials with fertilizers under the direction of a competent agronomist.* The least satisfactory arrangement is to make large numbers of analyses without adequate field controls.This is now being done in some Continental countries, and attractive lit1.le booklets are issued to each farmer, telling him how much phosphate, potash, etc., his soils contain, and how'much fertilizer he must apply to obtain satisfactory crops. The analysts plead, and no doubt correctly, that they have no time to make the necessary field experiments. In between comes the case, not uncommon in Britain, where the results are interpreted on a purely arbitrary scale based to some extent on field trials and still more or1 field observations.In advisory work,soil analysts in this country adopt the sound policy of putting the soils into categories, rather than giving figures to the farmers. Dr. Walker uses three categories: deficient, average, not deficient; other analysts use more categories: it is a question of finding the best way of conveying practical recommendations to farmers. For lime requirements, the advice is probably satisfactory, and reasonably so for phosphate; it may be less so for potash, especially in the wetter regions. During the war, advisoiy chemists in this country had the duty of deciding which fields were to be allowed additional potassic fertilizer. Thousands of soils had to be examined. In the drier English counties, where large acreages of potatoes and other potash-sensitive crops are grown, the allocations of extra potassium to soils in which analysis suggested a deficiency, played an important part in increasing crop yields.In the wetter Welsh counties the results were less certain, and chemists were not altogether happy .about them. t There can be no question, however, that the position in this country is improving: 5eld experiments of the 2 x 2 x 2 and 3 x 3 >: 3 NPK type are being extended, also trials involving differential manuring are being made, and at the same time the relevant soil analyses, Much valuable knowledge is thus accumulating. The results would gain in value and in interest if uniform analytical methods were used, as comparisons could then be made, not only within each province, but with soils in other provinces. But where methods are necessarily empirical, much latitude must be allowed. The comprehensive investigations of E. M. Crowther and his colleagues at Rothamsted on the manuring of sugar beet will, when completed, throw much light on this subject.$ We have obviously moved a long way from the ideas that formed the background of Dyer's early work to those current to-day. The general change has accorded with the change in science itself. The nineteenth century was a time of great generalisations, majestic in their comprehensiveness, magnificent in their .simplicity. So arose the idea of the simple division of soil phosphorus and potassium compounds into two groups, available and non- available. In this century the accumulation of awkward facts that could not be fitted into SO simple a scheme broke down the old generahations. Hall and Plymen pictured a great range of slightly differing compounds with no sharp dividing line. The modern idea is more complex still: the soil is viewed as the seat of many varied interactions involving colloids, micro-organisms, the plant roots and the weathering processes; with no sharp upper or lower * These were discontinued under the Hitler r6gime. f See G. W. Robinson, Chcm. and Ind., 1943, 171-74. 3 It is a pleasure to thank Dr. Crowther for much help given me in preparing this lecture.May, 19501 SUTTON AND MARKLAND : HORTVET THERMOMETERS 251 limits. We are in the stage of recognising the complexities, but not yet of expressing them in terms of mathematical equations and degrees of statistical significance as has been done for some other sciences. The soil analyst, however, has to do something: he is called on to advise the farmer and he must adopt methods that give useful answers. The methods are at present empirical, and that means that he must have close contact with field experiments to satisfy himself that they really are working well; he must have a mind sufficiently open to reject them when they cease to do so, and sufficiently alert to be able to find or devise others that will be better. His work lies largely in regions not yet tidied up by the science of the day; his equipment must always be a wide knowledge of scientific methods combined with ingenuity of invention, soundness of judgment and complete intellectual integrity. It is because Bernard Dyer possessed these qualities in so marked a degree that we honour his memory, and in the words of Ecclesiasticus, count him among those “that have left a name behind them, that their praises might be reported.” That, however, is no new situation for a consulting chemist.