ARSENIC 
 
PUBLIC 
 
 f\,r-. n - rt r\ ji rt n _n n 
 
 HEALTH 
 LIBRARY 
 
 UNIVERSITY OF CALIFORNIA. 
 
 Class 
 
 A 
 
+ it 
 
 *K* 
 
ARSENIC 
 
BY THE SAME AUTHOR 
 
 MILK- ANALYSIS : A Practical Treatise on the Ex- 
 amination of Milk and its Derivatives. Cream,. Butter, 
 and Cheese. By J. A. Wanklyn. Second Edition. 
 Crown Svo, 5s. 
 
 WATER-ANALYSIS : A Treatise on the Examination 
 of Potable Water. By J. A. WANKLYH and E T. Chap- 
 man. Tenth Edition, entirely rewritten. 12th Thousand. 
 Crown Svo, 5s. 
 
 AIR- ANALYSIS : A Practical Treatise. By J. A. 
 WAHKLYH and W. J. Cooper. With Appendix on 
 Illuminating Gas. Second Edition. Crown Svo, 5s. 
 
 BREAD-ANALYSIS: A Practical Treatise on the 
 Examination of Flour and Bread. By J. A. Wanklyn 
 and W. J. Cooper. Second Edition. Crown Svo, 5s. 
 
 SEWAGE ANALYSIS: A Practical Treatise on the 
 Examination of Sewage and Effluents from Sewage. By 
 J. A. Wanklyn and W. J. Cooper, including also a 
 Chapter on Utilisation and Purification. Crown Bvo, 
 7s. 6d. 
 
 LONDON 
 
 KEG AN PAUL, TRENCH, TRUBNER & CO., Ltd. 
 
 Paternoster House, Charing Cross Eoad, w.c. 
 
ARSENIC 
 
 BY 
 
 PROFESSOR J. ALFRED WANKLYN, M.R.C.S. 
 
 CORRESPONDING MEMBER OF THE ROYAL BAVARIAN ACADEMY OF 
 
 SCIENCES, 1869 
 
 HONORARY MEMBER OF THE UNIVERSITY OF EDINBURGH CHEMICAL SOCIETY 
 
 FORMERLY PUBLIC ANALYST FOR BUCKINGHAM, HIGH WYCOMBE, 
 
 PETERBOROUGH, AND SHREWSBURY 
 
 LECTURER ON CHEMISTRY AND PHYSICS AT ST. GEORGE'S HOSPITAL 
 
 PROFESSOR OF CHEMISTRY IN THE LONDON INSTITUTION 
 
 DEMONSTRATOR OF CHEMISTRY IN THE UNIVERSITY OF EDINBURGH, 1850 
 
 LONDON 
 KEGAN PAUL, TRENCH, TRUBNER & CO. L T i» 
 
 PATERNOSTER HOUSE, CHARING CROSS ROAD 
 1901 
 

 -^ 
 
 The rights of translation and of reproduction are reserved 
 
 Printed by Ballantyne, Hanson &> Co. 
 At the Ballantyne Press 
 
PREFACE 
 
 In the preface to a book like that which is now placed 
 before the English readers, a little personal detail is 
 not altogether uncalled for. The author is now an old 
 English chemist who has been recognised as a chemist 
 — at any rate in France and Germany — for forty-three 
 years. Eeference to such authorities as Poggendorff's 
 Biographical Dictionary, published about the year 
 1862, will disclose the name of the author, and the 
 English reader (if he should happen to be able to 
 read German) may learn how it came to pass that 
 the author of this book came to be looked upon as a 
 chemist in the year 1858. 
 
 The author was brought up to the medical profes- 
 sion, served an old-fashioned apprenticeship in Man- 
 chester, and duly presented himself for examination, 
 and passed the College of Surgeons in London in the 
 year 1856. At that date the author occupied the 
 humble, but very responsible position of private assis- 
 tant to Frankland, the well-known eminent chemist. 
 The popular excitement over the then recent Palmer 
 case was at its height, and accusations of having 
 
 V 
 
 ^990 
 
VJ PREFACE. 
 
 poisoned for the sake of the insurance money were 
 rife. One such case of suspected poisoning led to the 
 sending of the supposed poisoned remains to Frankland 
 at the Owens' College laboratory. Experience then 
 acquired turned the attention of the author to the 
 detection of poisons, and, in the fulness of time, after 
 the lapse of more than forty years, this book issues- 
 forth. 
 
 The Laboratory, New Malden, Surrey, 
 2$th February 1901. 
 
CONTENTS 
 
 PAGE 
 
 Preface v 
 
 Introduction i 
 
 ■5 
 26 
 29 
 35 
 
 CHAPTER I. 
 
 CHEMISTRY OF ARSENIC. 
 
 Section 1. — Arsenic in General, Arsenic in the Free or 
 
 Metallic State 
 
 Section 2. — Arsenious and Arsenic Acids and their Salts 
 Section 3. — The Chloride and the Hydride of Arsenic . 
 Section 4. — Organo-Metallic Compounds of Arsenic 
 Section 5. — Kakodylic Acid and its Salts 
 
 CHAPTER II. 
 
 DETECTION AND MEASUREMENT OF ARSENIC. 
 
 Section 1. — Marsh's Test ....... 39 
 
 Section 2. — Reinsch's Test 58 
 
 Section 3. — General Testing and Estimation of Arsenic . 68 
 
 Section 4. — Special Organo-Metallic Condition of Arsenic, <fec. 7 5 
 
 CHAPTER III. 
 
 THE MANCHESTER EPIDEMIC OF ARSENICAL POISONING AND 
 
 THE LESSONS TO BE LEARNT FROM IT . . . 8 1 
 
 vii 
 
ARSENIC, 
 
 INTKODUCTION. 
 
 This book has been called forth by the recent mys- 
 terious epidemic of chronic arsenical poisoning in 
 Manchester, Liverpool, and other towns in the North 
 of England . Mysterious the epidemic is justly called ; 
 inasmuch as, although a considerable proportion of 
 arsenic has been traced to the initial sulphuric acid, 
 very little arsenic has been detected in the sugar 
 arising from the action of the arsenical sulphuric 
 acid on starch, and still less arsenic in the poisonous 
 beer itself. 
 
 Indeed the analyses of the beer have disclosed 
 so very little arsenic that at first sight they would 
 almost seem to negative the idea of arsenical poisoning 
 by means of the beer. And one medical practitioner, 
 with special experience in the administration of arsenic 
 in skin diseases, has expressed himself as follows : — 
 
 " Facts of this kind give one reasonable ground 
 for wondering if the infinitesimally small amount of 
 arsenic found in impure beer can justly be held 
 
 1 A 
 
2 ARSENIC. 
 
 responsible for all the cases of peripheral neuritis at 
 present laid to its charge." * 
 
 The author of this book believes that the Man- 
 chester epidemic is chronic arsenical poisoning, and 
 that arsenic in sufficient quantity may have existed 
 in the beer, but that it existed in organo-mctallic com- 
 bination in a state in which it would escape recogni- 
 tion by the ordinary tests as applied by the analyst?. 
 
 Apart, however, from the circumstances of this 
 special outbreak many questions arise, which invest 
 the chemistry of arsenic with interest. 
 
 Arsenic in its commonest form — common arsenic, 
 as it is called — has, for generations, stood out as the 
 typical poison. And terrible is Orfila's description of 
 a typical case of acute arsenical poisoning. I quote 
 Orfila's description from a well-known text-book 
 published early in the last century. The poison in 
 acute cases begins to take effect in about a quarter of 
 an hour, and then — "An austere taste in the mouth; 
 frequent ptyalism ; continual spitting ; constriction of 
 the pharynx and oesophagus ; teeth set on edge ; hic- 
 cups ; nausea ; vomiting of brown or bloody matter : 
 anxiety ; frequent fainting fits ; burning heat at the 
 precordia ; inflammation of the lips, tongue, palate, 
 throat, stomach ; acute pain of stomach, rendering 
 the mildest drinks intolerable : black stools of an 
 
 * British Medical Journal 12th January 190 1, p. 85. See Paper on 
 the "Toleration of Arsenic/' by Robert W. Mackenna, M.A., M.I].. 
 Ch.B. Edin., Clinical Assistant to the Liverpool Skin Hospital. 
 
INTRODUCTION. 3 
 
 indescribable foetor ; pulse frequent, oppressed and 
 irregular, sometimes slow and unequal ; palpitation 
 of the heart ; syncope ; unextinguishable thirst ; burn- 
 ing sensation over the whole body, resembling a con- 
 suming fire ; at times an icy coldness ; difficult 
 respiration ; cold sweats ; scanty urine, of a red or 
 bloody appearance ; altered expression of counte- 
 nance ; a livid circle round the eyelids ; swelling and 
 itching of the whole body, which becomes covered 
 with livid spots, or with a miliary eruption ; pros- 
 tration of strength ; loss of feeling, especially in the 
 feet and hands; delirium, convulsions" . . .; "loss 
 of the hair, separation of the epidermis ; horrible con- 
 vulsions and death." The French was translated by 
 Dr. Andrew Ure, from whose book, published in 1824, 
 the above passage is quoted. 
 
 One of the most wonderful facts in connection with 
 the action of arsenic on the human subject is the toler- 
 ance of arsenic acquired by the Styrian peasantry. 
 The Styrian arsenic-eater is in the habit of taking 
 a dose of arsenic which, if administered to one of 
 ourselves, would bring on the frightful succession of 
 symptoms just described. 
 
 In Johnston's " Chemistry of Common Life," pub- 
 lished rather more than forty years ago, some particu- 
 lars of the Styrian arsenic-eating are brought forward. 
 In vol. ii., second edition, page 193, the following 
 may be read : — 
 
 " Arsenic is thus consumed chiefly for two purposes 
 
4 ARSENIC. 
 
 — First, to give plumpness to the figure, cleanness and 
 softness to the skin, and beauty and freshness to the 
 complexion. Secondly, To improve the breathing and 
 give longness of wind, so that steep and continuous 
 heights may be climbed without difficulty and exhaus- 
 tion of breath. Both these results are described as 
 following almost invariably from the prolonged use of 
 arsenic either by men or by animals." 
 
 Arsenic-eaters, as might be expected, commence the 
 practice of arsenic-eating by taking small doses of 
 common arsenic. Less than half-grain doses are taken 
 to begin with, two or three times in the week. The 
 dose is very gradually and cautiously increased until it 
 reaches about two grains ; and this practice of arsenic- 
 eating may be continued for forty years without appa- 
 rent detriment to the system. 
 
 Recent medical experience points to arsenic as one 
 of the most potent medicinal agencies at the disposal 
 of the physician. That it is a specific in cases of 
 phthisis has been asserted, and may be true to some 
 extent. 
 
 The form of arsenic recommended for employment 
 for this purpose is Bunsen's kakodylic acid, the mar- 
 vellous substance containing rather more than half its 
 weight of metallic arsenic and yet, notwithstanding 
 that it is very soluble in water, absolutely non- 
 poisonous. More than sixty years have elapsed since 
 Bunsen's discovery of this wonderful compound of 
 arsenic, and it is only within the last few years that 
 
INTRODUCTION. 5 
 
 the attention of the medical profession has been called 
 to the possibilities of a remedy of this description. 
 
 In the proper place, later on in this book, a very de- 
 tailed description of this substance will be given, and 
 its various chemical relationships set forth. 
 
 Under certain conditions, which are possible and by 
 no means unlikely of realisation in the human body, 
 chemical changes might be set up which would trans- 
 form kakodylic acid into a deadly poison. 
 
 The reflection arises that there is pressing need for 
 diligent study of the chemistry of arsenic, if substances 
 like kakodylic acid are to come into use in medicine. 
 
CHAPTER I. 
 
 CHEMISTRY OF ARSENIC. 
 
 Section 1. — Arsenic in general, Arsenic in the free 
 or metallic state. 
 
 Arsenic in the form of the sulphnret and the oxide, 
 which is the arsenic of commerce, was known in very 
 ancient times, and is said to have been mentioned by 
 Avicenna in the eleventh century, and received its 
 name in token of its poisonous nature. 
 
 When the common arsenic of commerce was first 
 reduced so as to yield its metallic constituent is not on 
 record ; but to Brandt, in the year 1733, is due the 
 credit of carrying out the earliest methodical investiga- 
 tion into the chemical character of arsenical substances, 
 and showing that they contain a peculiar metallic sub- 
 stance existing in all of them in a state of chemical 
 combination. 
 
 In the year 1755 the Swedish chemist, Scheele, 
 discovered arsenic acid and the gas, arseniuretted 
 hydrogen. 
 
 Proust made fairly accurate analyses of the two 
 oxides of arsenic — analyses only differing by a few 
 
8 ARSENIC. 
 
 tenths per cent, from that which is, at the present clay, 
 accepted as the correct figures — which are duly re- 
 corded in publications dating within a few years of 
 the year 1 800 ; showing, indeed, that certain arsenical 
 compounds afforded some of the earliest instances of 
 fair quantitative accuracy in chemistry. 
 
 The atomic weight of arsenic was determined by 
 Berzelius, and also — and quite independently — by 
 Pelouze : and the figure 75.0 is accepted by chemists. 
 The symbol for arsenic is As = 75. 
 
 In the older books, issued at the beginning of the 
 nineteenth century, arsenic is spoken of as one of the 
 semi-metals, and that designation holds good at the 
 present day. In the elementary state arsenic possesses 
 one of the characteristics of metals, in a high degree, 
 viz. metallic lustre, but the general character of its 
 compounds is that rather of a non-metal. In its com- 
 pounds it resembles phosphorus more closely perhaps 
 than any other element. It is a triad and a pentad ; 
 and, as will be related later on, it possesses pre-emi- 
 nently the power of entering into intimate organic 
 combination with carbo-hydrogens. 
 
 Arsenical compounds, at the ordinary temperature, 
 do not undergo reduction so as to yield elementary, or 
 metallic arsenic ; but, at a low red heat and at higher 
 temperatures, arsenical compounds are very prone to 
 undergo reduction to the metallic state. 
 
 Soine arsenical compounds (arseniuretted hydrogen 
 for instance) simply fall to pieces when they are heated 
 
CHEMISTRY OF ARSENIC. 9 
 
 to low redness, and as the temperature is lowered, 
 metallic arsenic is deposited in a solid state. Other 
 arsenical compounds require the presence of some 
 appropriate reducing agent in order that the arsenic 
 may assume the metallic state when they are heated. 
 Some again may need, as a preliminary, a roasting in 
 air ; but almost all arsenical compounds evolve metallic 
 arsenic by heating to redness in a reducing atmosphere, 
 or in an atmosphere alternately oxidising and reducing. 
 The extreme ease with which arsenic is reduced has 
 almost become proverbial among chemists, and forms 
 one of the most striking analytical characters of that 
 substance. 
 
 The production of arsenic on the large scale is thus 
 described in a well-known modern text-book. Native 
 arsenide of iron, or arsenical pyrites FeAsS contained 
 in earthen retorts or tubes, is heated in a furnace ; the 
 whole of the arsenic then distils off, leaving a residue 
 of iron or of sulphuret of iron in the retorts. The 
 retorts are laid horizontally in the furnace, and tubes 
 made out of thiu iron plates rolled up into the form of 
 a tube are inserted into the mouths of the retorts and 
 an earthenware receiver is luted on. The arsenic 
 condenses chiefly in the iron tube, arid when the tube 
 has cooled is detached by unrolling the tube. By 
 operating in that manner the arsenic is obtained in 
 the form of a compact mass which is nearly white. 
 
 At Altenberg in Silesia metallic arsenic is obtained 
 by heating arsenious acid with charcoal in earthen 
 
IO ARSENIC. 
 
 crucibles covered by conical iron caps or by earthen 
 inverted crucibles ; but, though the yield is larger, the 
 quality is not so good as in the first described pro- 
 cess, the arsenic being pulverulent aud grey. 
 
 Sulphuret of arsenic may be reduced to metallic 
 arsenic by heating it with black flux (a mixture of two 
 parts of cream of tartar and one of nitre previously 
 ignited) in a crucible covered with an inverted crucible 
 luted to it with clay and sand. The arsenic forms a 
 sublimate on the inside of the inverted crucible. 
 
 Metallic arsenic passes, without first melting, into 
 the form of vapour at a temperature not very far re- 
 moved from the boiling-point of mercury. The arseni- 
 cal vapour condenses on cooling and yields a solid, the 
 appearance of which varies according to the physical 
 conditions to which it is subjected. During the melt- 
 ing the metal emits a disagreeable smell like garlic, 
 which is said to be very characteristic. The colour of 
 metallic arsenic has been variously described. Steel- 
 grey says one modern author ; bluish-white, the colour 
 subject to tarnish, becoming yellowish and ultimately 
 black on exposure to the air, says an old author. 
 
 Metallic arsenic is very brittle and may be easily 
 powdered. It crystallises in rhombohedrons. 
 
 It is lighter than antimony, its specific gravity being 
 between 5.62 and 5.96. 
 
 It is a little heavier than selenium, but not quite so 
 heavy as tellurium. 
 
 Under appropriate conditions metallic arsenic ex- 
 
CHEMISTRY OF ARSENIC. II 
 
 hibits great chemical activity. When it is volatilised 
 it yields a vapour which burns with a faint blue flame, 
 producing As 2 3 , arsenious acid. Powdered metallic 
 arsenic takes fire spontaneously in chlorine gas, pro- 
 ducing As01 3 , the only known chloride of arsenic. 
 
 A mixture of one part of powdered metallic arsenic 
 with three parts of powdered chlorate of potash forms 
 a very dangerous explosive, which explodes most vio- 
 lently when struck with a hammer on an anvil. Such 
 a mixture is dangerous to make, and should be made 
 only on a very small scale, and with avoidance of 
 friction. 
 
 Nitrates likewise make explosive mixtures with 
 arsenic. 
 
 Metallic arsenic has no action on water or steam, 
 but appears to combine with any oxygen that may be 
 held in aqueous solution. 
 
 Dilute acids have little or no action, but boiling 
 hydrochloric acid appears to dissolve it slowly (pro- 
 bably indirectly through the action of oxygen taken 
 up from the atmosphere). 
 
 Concentrated sulphuric acid has no action in the 
 cold, but is reduced to sulphurous acid and even to 
 the condition of free sulphur on being boiled with 
 metallic arsenic. 
 
 Nitric acid attacks metallic arsenic, yielding ulti- 
 mately arsenic acid. 
 
 The common oils and fats are attacked by metallic 
 arsenic when they are somewhat strongly heated with 
 
12 ARSENIC. 
 
 it, giving a black liquid which becomes semi-solid on 
 cooling. 
 
 When sulphur is melted with arsenic combination 
 takes place, and various sulphurets are produced 
 according to the proportions employed. 
 
 Almost every metal combines with metallic arsenic 
 under the influence of heat, and forms compounds 
 which are generally called alloys, some, however, being 
 dignified as arseniurets or arsenides. 
 
 The general effect of arsenic upon a metal is to 
 render it brittle and to lower its melting-point. The 
 colour is often altered and whitened. 
 
 Mercury takes up much arsenic when it is boiled with 
 arsenic, and forms a grey amalgam consisting of five 
 parts of mercury and one part of arsenic. 
 
 Gold takes up one-sixtieth of its weight of arsenic, 
 giving an alloy pale in colour, and brittle and much 
 harder than gold. 
 
 Platinum alloys itself with arsenic, and the alloy of 
 platinum with arsenic has played a useful part in the 
 metallurgy of platinum. For that purpose, however, 
 it is arsenious acid rather than metallic arsenic which 
 is used. 
 
 Molten silver takes up as much as one-fourteenth 
 of arsenic, forming a brittle yellow alloy. 
 
 The alloy of arsenic with copper, which is known as 
 white copper or white tombac, has the composition 
 Cu 2 As, and may be prepared by heating together 
 finely divided copper and arsenic. 
 
CHEMISTRY OF ARSENIC. 1 3 
 
 An arseniuret or arsenide of iron having the 
 composition FeAs has been obtained by heating 
 together in a closed crucible a mixture of 54 parts 
 of iron with 108 parts of metallic arsenic. This alloy 
 was white and brittle, easily reduced to powder. It 
 contained : — 
 
 Arsenic . . . 75 63 
 
 Iron 56 37 
 
 131 100 
 
 Bismuth, antimony, tin, zinc, and lead, each of these 
 metals combines with arsenic, when metallic arsenic 
 and the metal are heated together in closed vessels. 
 
 Heated with potassium or sodium metallic arsenic 
 forms alloys which possess the property of decom- 
 posing water, acidulated if necessary, and yielding 
 arseniuretted hydrogen by that decomposition. 
 
 Such alloys if they were heated in a current of dry 
 air or oxygen would combine with the oxygen, at 
 any rate, at temperatures bordering on redness ; and 
 arseniates would be the result. 
 
 The alloys of some of the heavy metals with 
 arsenic are difficult of complete decomposition by 
 heat. When heated to redness in the air they some- 
 times throw off part of the arsenic in the form of 
 arsenious acid, but a portion of the arsenic remains 
 behind in the form of an arseniate. By adopting 
 the plan of fusing with nitre, the alloys of arsenic 
 with heavy metals may be converted into a mixture 
 
14 ABSENIC. 
 
 of arseniate of potash (which will dissolve in water 
 on subsequent cooling), and oxide of the heavy metal, 
 which will remain insoluble, aud be separable from 
 the alkaline arseniate in solution. By that means 
 the analytical difficulty 'presented by some of these 
 alloys may be surmounted. 
 
 Metallic arsenic has no tendency whatever to unite 
 directly with carbon, although, as will be explained 
 in the proper place, it has an especial and peculiar 
 aptitude for entering into intimate organic relation 
 with hydrocarbon radicals when the proper conditions 
 are complied with. It is, however, worthy of mention 
 that a case is on record where the charcoal furnished 
 by an organic nitrogenous substance retained arsenic 
 so obstinately that the existence of arsenic in the 
 original organic compound was, for a long time, 
 overlooked. Such an example of the concealment of 
 the presence of arsenic was furnished by arseniate of 
 rosaniline, the nature of which was not at first re- 
 cognised. In that instance it was necessary to burn 
 up the charcoal in a current of air or oxvgen in 
 order to obtain the arsenic in the form of arsenious 
 acid. 
 
 Metallic arsenic is found in Nature in several 
 mining districts in Saxony Bohemia, and Transyl- 
 vania; also in Norway and the United States of 
 America. 
 
 It occurs much more abundantly in a state of com- 
 bination with various metals and with sulphur. Its 
 
CHEMISTRY OF ARSENIC. I 5 
 
 commonest compounds are its combinations with iron, 
 copper, nickel, cobalt, and with sulphur. There is no 
 mining district in which arsenic is unknown, but it 
 is especially abundant in Germany and in SpaiD, in 
 the province of Catalonia. It is found in Cornwall 
 in tolerable abundance. Pyrites contain more or less 
 of it, and native sulphur is never quite free from it. 
 Nearly every iron ore contains it in some shape or 
 other, though the proportion is sometimes very small. 
 
 The actual proportion of arsenic in the earth's crust 
 is no doubt very small, but traces of arsenic are found 
 almost everywhere on the surface of this planet. In 
 the mining districts it is discharged into the atmos- 
 phere by many metallurgical operations, even in the 
 daily life in town and country it finds its way into 
 the atmosphere, since wherever coal is burnt traces 
 of arsenious acid are sent into the air. Every shower 
 of rain washes arsenic out of the air and distributes 
 it over the land or into the sea. Minute traces of 
 arsenic are all-pervading, and, as a necessary conse- 
 quence, the mere detection of arsenic is devoid of 
 meaning unless it is, to some extent, a quantitative 
 operation. 
 
 Sect ion 2. — Arsenious and Arsenic Acids and 
 their Salts. 
 
 There are two oxides of arsenic, viz. arsenious acid 
 and arsenic acid. The first of these ask 
 
1 6 ARSENIC. 
 
 "Arsenic " of commerce, white arsenic, as it is called. 
 It is the volatile product of the roasting of the various 
 ores of arsenic, and its production has been described 
 by an old writer in these words : " The ore is thrown 
 into a furnace resembling a baker's oven, with a flue 
 or horizontal chimney nearly two hundred yards long, 
 into which the fumes pass, and are condensed into 
 a greyish or blackish powder. This is refined by a 
 second sublimation in enclosed vessels with a little 
 potash to detain impurities." The finished product 
 is a heavy white solid with the following percentage 
 composition : — 
 
 Arsenic . . . . 75-76 
 
 Oxygen . . . . .24.24 
 
 100.00 
 
 Its chemical formula is As 9 3 , and it is a mixture of 
 the two or three allotropic varieties of arsenious acid. 
 There is the vitreous variety, which is amorphous and 
 quite distinct from the crystalline variety ; but the 
 different varieties pass into one another with great 
 facility, and whether one or another form predomi- 
 nates in a mass of arsenious acid depends upon slight 
 alterations in the physical conditions. Arsenious acid 
 volatilises at about 21 8° Centigrade, giving a colourless 
 vapour which, if it be allowed to condense very gradu- 
 ally at a temperature only a few degrees below 21 8°, 
 passes through a condition of partial fusion before it 
 assumes the completely solid state, and under these 
 
CHEMISTRY OF AKSENIC. 1 7 
 
 physical conditions becomes the vitreous or amorphous 
 variety of arsenious acid. 
 
 The crystalline variety is obtained by allowing the 
 vapours to condense in contact with surfaces cooled 
 much below 21 8°, under conditions where there can 
 be no partial fusion of the sublimate. 
 
 The amorphous arsenious acid is much heavier than 
 the crystalline variety; its specific gravity is 3.7385, 
 whilst the crystals have a specific gravity of 2.695. 
 
 The solubility of the different arsenious acids varies 
 considerably, especially in cold water. A boiling satu- 
 rated solution of arsenious acid contains some eight or 
 ten grammes of arsenious acid in 100 c. c. of the solu- 
 tion, and on cooling deposits crystals of arsenious acid, 
 leaving a cold saturated solution containing 3.3 grammes 
 in the 100 c. c. of cold solution. But when the attempt 
 is made to obtain a cold saturated solution without 
 a preliminary preparation of the hot solution, when 
 excess of finely powdered arsenious acid is persistently 
 shaken up with cold water, then only one-tenth of 
 that quantity of As 2 3 passes into solution ; and under 
 those conditions only 0.3 gramme of As 2 3 is found 
 in 100 c. c. of the cold solution. 
 
 The following experiment is very instructive. Take 
 one gramme of finely powdered As 2 3 and shake it up 
 with one litre of cold distilled water, and note how 
 slowly the substance dissolves, and that in no reason- 
 able time, and that with no reasonable persistence in 
 shaking up is it possible to induce the whole of the 
 
 B 
 
1 8 ARSENIC. 
 
 gramme or nearly the whole of the gramme to enter 
 into solution. The results of such experiments appear 
 to indicate that there is a variety of As 2 3 which is 
 almost insoluble in cold water. 
 
 In boiling dilute hydrochloric acid As 2 3 dissolves 
 more readily than in boiling water, and on cooling 
 As 2 3 separates out of solution uncombined either 
 with water or with HC1. 
 
 This property of separating out of solution in the 
 anhydrous condition is one of the points of distinction 
 between arsenious acid and the other oxide of arsenic, 
 viz. arsenic acid, which enters into real combination 
 with water producing hydrates, and can only be de- 
 prived of its combined water by the employment of 
 considerable heat. 
 
 Sulphuretted hydrogen gas transforms As 2 3 into 
 the yellow sulphuret of arsenic As 9 S 3 thus : — - 
 
 3H 2 S + A S . 2 3 = 3 H 2 0+A S2 S 3 . 
 
 In order that the separation of the yellow sulphuret 
 may be complete it is advisable to make an addition 
 of an excess of dilute hydrochloric acid, and under 
 that condition the precipitation of the sulphuret of 
 arsenic is absolute. The yellow sulphuret is constant 
 in composition, and is one of the most characteristic 
 compounds of arsenic. It is one of the forms in which 
 arsenic is weighed in performing a quantitative esti- 
 mation of arsenic. It is the sulphur representative 
 of arsenious acid, and dissolves with great ease in 
 
CHEMISTRY OF ARSENIC. 1 9 
 
 ammonia, or carbonate of ammonia, or sulphuret of 
 ammonia, from which solutions it may be reprecipi- 
 tated by the addition of hydrochloric acid. 
 
 Arsenious acid is one of the weakest acids. Its 
 aqueous solution reddens litmus feebly, but on the 
 other hand, although it dissolves freely in solutions 
 of alkaline carbonates, it does not drive out the 
 carbonic acid at ordinary temperatures. It dissolves 
 in excess of caustic potash or soda, but it does not 
 neutralise the alkali, and even the carbonic acid of the 
 air drives out the arsenious acid, which in course of 
 time separates out in the form of crystals. 
 
 In solution of ammonia it dissolves more readily 
 than in water devoid of ammonia, but the ammonia 
 passes off when the water evaporates, and nothing but 
 As 2 3 remains behind as a residue. Excess of the 
 acid forms soluble acid salts with lime, strontia, or 
 baryta, and solutions of these acid-salts give basic 
 precipitates when lime-water, strontia-water, or baryta- 
 water is added in excess. 
 
 The arsenite of copper 2CuO,As 2 3 (Scheele's 
 green) is a very well-known substance. It is formed 
 on mixing together an aqueous solution of arsenious 
 acid with solution of ammonio-sulphate of copper, when 
 a brilliant yellow green precipitate falls down. This 
 precipitate is very soluble either in ammonia or acids ; 
 and, of course, would only make its appearance in the 
 absence of excess of ammonia or acid. 
 
 The arsenite of silver 2Ag 2 0,As 2 3 forms a very 
 
20 AESENIC. 
 
 characteristic precipitate when ammonio-nitrate of silver 
 and the acid are brought together. The colour of this 
 silver precipitate is canary yellow. Like the copper 
 precipitate just mentioned it is very sensitive to excess 
 of ammonia or of acid, which has to be carefully 
 guarded against in order to avoid failure. 
 
 Recently precipitated hydrated peroxide of iron, 
 when added in great excess, possesses the property 
 of removing arsenious acid from its aqueous solution 
 with great completeness. For this purpose the amount 
 of the iron precipitate is required to be ten times the 
 weight of the arsenious acid ; and in this calculation 
 the iron is to be reckoned as anhydrous peroxide. So 
 thorough is the action, and so little objectionable is 
 the agent, that Bunsen and Berthold have recom- 
 mended the use of it as an antidote in cases of arseni- 
 cal poisoning when the poisoning is quite recent. One 
 of the best forms of the remedy is the use of a strong 
 solution of ferric chloride mixed with calcined magnesia 
 in excess. 
 
 Arsenious acid is interesting inasmuch as it is one 
 of the most stable forms of arsenic. As has been men- 
 tioned, this is the form which arsenic assumes when 
 the ores of arsenic are roasted in contact with the 
 atmosphere. It is also the only form which arsenic 
 takes when metallic arsenic is heated in excess of air. 
 Even the other oxide of arsenic, viz. As 2 5 , arsenic 
 acid, is resolved into As 2 3 and free oxygen when a 
 high red heat is reached. The conditions being — pre- 
 
CHEMISTRY OF ARSENIC. 21 
 
 sence of excess of oxygen, high red heat, and absence of 
 alkali, or powerful basic substances, the sole form taken 
 by arsenic is As 2 3 , arsenious acid. 
 
 When, however, the last condition is not complied 
 with, and when there is excess of alkali or powerful 
 basic substance, then the stable form is the other oxide 
 of arsenic, viz. As 2 5 arsenic acid, which in that case 
 enters into combination and yields an arseniate of the 
 alkali or strongly basic substance. Under these con- 
 ditions arsenites take up oxygen and become arseniates. 
 And, when there is no free oxygen to be taken up, then 
 the arsenious acid, so to speak, feeds upon itself, and 
 five molecules of it are resolved into three molecules of 
 arsenic acid and four atoms of metallic arsenic, accord- 
 ing to this equation : — 
 
 5As,0 3 = 3 As 2 5 + As 4 . 
 
 "Dr. Wollaston first observed, that when a mixture 
 of arsenious acid with quicklime is heated in a glass 
 tube, at a certain temperature ignition suddenly per- 
 vades the mass and metallic arsenic sublimes." 
 
 If then some of the arsenious acid in contact with 
 strong bases, at a high red heat, acts towards itself 
 as a reducing agent and evolves metallic arsenic, one 
 would anticipate that the presence of a highly re- 
 ducing agent, such as charcoal, should bring about 
 complete reduction of the whole of the arsenic, and 
 that expectation is thoroughly borne out by experi- 
 ment. And one of the most characteristic reactions of 
 arsenious acid is the production of the arsenical mirror 
 
22 ARSENIC. 
 
 by reduction of that substance by red-hot charcoal, 
 which is so placed in a narrow tube of hard glass that 
 the volatilised arsenious acid is compelled to come into 
 contact with it and undergo reduction, and deposit a 
 shining ring of metallic arsenic in the cooler part of 
 the tube. 
 
 Arsenic acid, sometimes called arsenic oxide, or 
 arsenic pentoxide, or arsenic anhydride, is the only 
 other compound of arsenic with oxygen. Its formula 
 is As 2 5 . It cannot be obtained from arsenious 
 acid by simply heating that substance in contact with 
 atmospheric air or oxygen. At very elevated tem- 
 peratures arsenic acid is indeed split up into arsenious 
 acid and free oxygen. Arsenic acid is obtainable from 
 arsenious acid by several roundabout operations. 
 
 As has been mentioned the admixture of quick- 
 lime with arsenious acid brings about a very remark- 
 able decomposition when the materials are strongly 
 heated ; and under those conditions 60 per cent, of 
 the arsenic in the arsenious acid passes into the con- 
 dition of the higher acid of arsenic, and 40 per 
 cent, of the arsenic sublimes in the condition of 
 metallic arsenics. In this instance the process is a 
 furnace operation, and the acid forms a compound 
 with the lime. 
 
 The usual method resorted to in order to convert 
 arsenious into arsenic acid is to boil it with nitric 
 acid or with a mixture of nitric and hydrochloric acids. 
 In that manner an aqueous solution of the hydrate 
 
CHEMISTRY OF ARSENIC. 23 
 
 of arsenic acid is produced. By evaporating by the 
 application of heat, the water is driven off and there 
 remains a white concrete substance differing in many 
 respects from arsenious acid. This residue is arsenic 
 acid. Its specific gravity is 3.391. It is not crystal- 
 line. It is fixed in the fire (at anyrate at a low red 
 heat), whereas arsenious acid volatilises at about 21 8° 
 Centigrade. Its composition is : — 
 
 Arsenic . . . . .65.2 
 Oxygen 34.8 
 
 100. o 
 
 If the cooled mass be placed in contact with water 
 in excess it will enter into chemical composition with 
 water, giving the tri-hydrate As 2 5 ,3H 2 0, which is 
 sometimes written thus, H 3 As0 4 , the formula being- 
 halved. 
 
 The aqueous solution of this hydrate is very strongly 
 acid, and is strikingly different from the aqueous 
 solution of arsenious acid. 
 
 There is a very close and far-reaching resemblance 
 between arsenic and phosphoric acid. Just as there 
 are three different hydrates of phosphoric acid, so 
 there are three corresponding hydrates of arsenic 
 acid ; but the hydrates of arsenic acid pass more 
 readily into each other than in the case of phosphoric 
 acid. And simple solution in water resolves both of 
 the other hydrates of arsenic acid into the tri-hydrate. 
 
 The salts formed by arsenic acid with the alkalis 
 
24 ARSENIC. 
 
 are soluble in water. The salts of baryta, strontia, 
 and lime are all of them soluble in excess of acid. 
 The acid salts are soluble but the basic salts are 
 insoluble in water. In presence of ammoniacal salts 
 the basic salts of baryta, strontia, and lime become 
 soluble. 
 
 There is an ammonia arseniate of magnesia which 
 "is a very close representative of the corresponding 
 phosphate, and which is admirably adapted for quan- 
 titative employment. Its formula is As 2 5 (MgO) 9 
 (NH 4 ) 2 + 1 2H 2 0. It is crystalline, very insoluble 
 in strongly ammoniacal liquids, just as the corre- 
 sponding phosphate. 
 
 The arseniate of silver, which is brick red in colour, 
 is a very characteristic arsenical compound. It is 
 obtained by mixing solution of arsenic acid with 
 ammonio-nitrate of silver. For this purpose the 
 ammonio-nitrate of silver must be very carefully pre- 
 pared — only just enough ammonia being added to 
 barely dissolve the precipitate given on the first 
 addition of drops of ammonia to the nitrate. Excess 
 either of acid or of ammonia redissolves this precipi- 
 tate. The formula of the precipitate is Ag 3 As0 4 . 
 
 There is also a preci|)itate when solutions of sulphate 
 of copper and arsenic acid are brought together and 
 very carefully dosed with very dilute ammonia. This 
 precipitate has the formula HCuAs0 4 and is blue- 
 green in colour, and is soluble in slight excess, either 
 of ammonia or nitric acid. 
 
CHEMISTRY OF ARSENIC. 25 
 
 Just as phosphoric acid forms a most insoluble 
 yellow precipitate with molybdate of ammonia in 
 presence of much nitric acid and nitrate of ammonia, 
 so likewise does arsenic acid. 
 
 This test is especially applicable when very small 
 quantities of arsenic acid or its salts are required to 
 be tested for the presence of arsenic acid. 
 
 It is essential that the molybdate of ammonia be 
 added in excess, and that excess of nitric acid be 
 present, and the mixture should be warmed. These 
 conditions having been complied with the presence of 
 arsenic acid, or its salts, will be announced by the 
 appearance of a yellow colour in the solution, and, if 
 more than a trace of arsenic acid be there, a yellow 
 precipitate of the arsenio-molybdate of ammonia will, 
 by-and-by, settle down. Of course it will be under- 
 stood that a testing, of this kind, for arsenic acid is 
 valid only in the absence of phosphoric acid. 
 
 The appropriate solution of molybdate of ammonia 
 is prepared by dissolving one part of molybdic acid in 
 eight parts of strong ammonia, and then adding twenty 
 parts of nitric acid. Forty times as much molybdic 
 acid must be used in order to precipitate a given 
 quantity of arsenic acid. 
 
 The behaviour of arsenic acid and the salts of arsenic 
 acid towards sulphuretted hydrogen differs somewhat 
 from that of arsenious acid towards that reagent. The 
 solution of arsenic acid, or arseniate having been acidi- 
 fied by means of dilute hydrochloric acid, sulphuretted 
 
26 ARSENIC. 
 
 hydrogen gas may be passed into the liquid, when the 
 following may be noted. In the cold, at ordinary 
 temperatures, the action of the gas is exceedingly slow, 
 twenty-four hours being required to complete the 
 production of the yellow sulphuret As 2 S 3 ; but, if the 
 liquid be warmed up to 6o° or 70° Centigrade, then 
 the production of the yellow sulphuret is almost 
 immediate. 
 
 The equation expressing the chemical reaction is : — 
 5H 2 S + As 2 5 = 5H 2 + As 2 S 3 + S 2 , 
 and, as will be perceived, there is production of sulphur 
 in addition to the yellow sulphuret As 2 S 3 . 
 
 Instead of directly passing the sulphuretted hydrogen 
 through a heated solution of arsenic acid it will be 
 found convenient to employ sulphurous acid, and, in 
 that way, reduce the arsenic acid into arsenious acid 
 before resorting to sulphuretted hydrogen. The 
 chemical change may be expressed thus : — 
 As 2 5 + 2S0 2 = As 2 3 + 2S0 3 . 
 Of course the 2S0 3 acts upon water and becomes 
 2(S0 3 H 9 0). The solution should be warmed in order 
 to facilitate the chemical change. 
 
 Section 3. — The Chloride and the Hydride of 
 
 Arsenic. 
 
 Chloride of arsenic, formula AsCl 3 , theoretical 
 
 vapour-density = 6. 289. Dumas found 6.300. It is 
 
 a heavy colourless liquid, which decomposes water 
 
 when brought into contact with water. Its boiling- 
 
CHEMISTRY OF AltSENIC. 2 J 
 
 point is 1 32° Centigrade. It is produced by distilling a 
 mixture of common salt and arsenious acid with excess 
 of sulphuric acid, and is a highly poisonous liquid which 
 fumes in moist air owing to the action of its vapour on 
 moisture. There is no pentachloride of arsenic. 
 
 Bromine and iodine form corresponding compounds. 
 The iodide Asl 3 is a beautiful substance described as an 
 orange-coloured sublimate. It is prepared by heating 
 a mixture of iodine and metallic arsenic. It is very 
 soluble in hot water, and, curiously enough, such a 
 solution when boiled down to dryness leaves a residue 
 of iodide of arsenic. 
 
 The hydride of arsenic, also called arseniuretted 
 hydrogen, AsH 3 , is a colourless gas, produced in carrying 
 out the celebrated Marsh test for arsenic, but, in that 
 instance, it is mixed with a large volume of hydrogen 
 gas. It has not often been prepared in a state of even 
 approximate purity. In anything like a state of 
 purity it is terribly poisonous — said to be the most 
 poisonous substance known. 
 
 Its vapour-density, according to Dumas, is 2.695, 
 which agrees very closely with the usual formula 
 assigned to it. The extremely poisonous nature of 
 this compound renders it very dangerous, and has no 
 doubt hindered investigation. 
 
 AsH 3 is the chemical representative of ammonia 
 NH 3 , and on that account very interesting to the 
 chemist. In a state of purity — unaccompanied by 
 hydrogen — it was obtained, by Soubeiran, by the 
 
28 ARSENIC. 
 
 action of strong hydrochloric acid upon an alloy of 
 arsenic and zinc, in which the arsenic was in excess 
 over the zinc. Such an alloy Soubeiran procured by 
 fusing a mixture of equal weights of arsenic and 
 granulated zinc ; and. using that alloy, Soubeiran 
 obtained arseniuretted hydrogen which dissolved in 
 a saturated aqueous solution of sulphate of copper, 
 and left no residue of unabsorbed hydrogen. The 
 gas has a most disagreeable smell, like garlic. Unlike 
 ammonia, which dissolves in water to an enormous 
 extent, arseniuretted hydrogen is only slightly soluble 
 in water; ioo volumes of water dissolving only about 
 20 volumes of AsH 3 . Furthermore, it should be borne 
 in mind that arseniuretted hydrogen has no tendency 
 to combine with hydrochloric acid — in that respect, 
 too, being utterly different from ammonia. It is like 
 ammonia in its' passiveness in relation to the alkalis. 
 
 Its chemical activity is that of a most powerful and 
 most peculiar reducing agent, and the action is, as a 
 rale, rapid and apparently instantaneous. The common 
 solutions of gold, platinum, and silver are. at once, 
 decomposed by this gas, the metal being precipitated 
 whilst the arsenic in the arseniuretted hydrogen suffers 
 oxidation and enters iuto solution in place of the 
 noble metal which is assumiuo- the metallic state. 
 
 With solution of sulphate of copper there is, as has 
 been mentioned, a very peculiar action, and a preci- 
 pitate is formed ; but the precipitate does not consist of 
 metallic copper but of an alloy of arsenic with copper. 
 
CHEMISTKY OF ARSENIC. 29 
 
 With solution of acetate of lead acidulated with 
 acetic acid there is no visible action on bringing the 
 gas into contact with it : which distinguishes the gas 
 from sulphuretted hydrogen. 
 
 Section 4. — Or gano- Metallic Compounds of Arsenic. 
 
 In the year 1760 a very extraordinary liquid pos- 
 sessing the property of taking fire spontaneously on 
 exposure to the air was discovered by Cadet. The 
 liquid became known as Cadet's fuming liquor or 
 alcarsin, and the study of it by the great German 
 chemist, Bunsen, rather more than seventy years 
 later, laid the foundation of much that is embraced 
 by modern organic chemistry. 
 
 Cadet's fuming liquor is produced by the careful 
 distillation, at moderately high temperatures, in a 
 retort, of an intimate mixture of powdered arsenious 
 acid and dry acetate of potash. 
 
 In that experiment, instead of a simple reduction 
 of the arsenious acid so as to yield a sublimate of 
 metallic arsenic, there was also produced a new and 
 strange organic liquid containing arsenic in intimate 
 union with the hydro-carbon radical methyl. Bunsen's 
 work established the real nature of Cadet's fuming 
 liquor ; and, inasmuch as this compound of arsenic 
 with methyl was the first recognised compound of 
 any metal or semi-metal with any hydro-carbon radical, 
 Bunsen is justly held to be the discoverer of this 
 
30 
 
 ARSENIC. 
 
 class of chemical substances. Organo-metallic com- 
 pounds,, as possible existences, were discovered when 
 Bunsen recognised kakodyl in Cadet's fuming liquor. 
 At the time of Bunsen's great investigation (be- 
 ginning with the year 1837 and ending with the 
 year 1842) the now familiar notion that acetates 
 contain the organic radical methyl, as an integral 
 part of the acetic acid, had not presented itself to 
 the minds of chemists, and the explanation of the 
 production of such a substance as Bunsen's kakodyl by 
 such an operation as the action of arsenious acid on 
 acetate of potash was much more difficult than it is 
 in the light of to-day's chemistry. 
 
 The chemical change taking place when arsenious 
 acid is distilled in contact with dry acetate of potash 
 may be formulated thus : — 
 
 4 Molecules of 
 Acetate of Potash. 
 
 KOCOCH 8 
 KOCOCHg 
 KOCOCH s 
 KOCOCH, 
 
 One Molecule of 
 Arsenious Acid. 
 
 As.,0« 
 
 KOCOO 
 
 ■f 
 
 KOCO 
 
 
 KOCOO 
 
 
 KOCO 
 
 
 Neutral Carbonate 
 of Potash. 
 
 Carbonic 
 Acid. 
 
 2K 2 C0 3 + 
 
 2C0 2 
 
 Oxide of Kakodyl. 
 
 As 2 (CH 3 ) 4 
 
 Alkarsin or Oxide 
 of Kakodyl. 
 
 As 2 (CH s ) 4 0. 
 
CHEMISTRY OF ARSENIC. 3 1 
 
 This equation requires that 392.4 parts of acetate of 
 potash should be taken along with 198 parts arsenious 
 acid, being about double as much potash salt as 
 arsenious acid. Bunsen used equal proportions of 
 acetate and the acid ; that is to say, the acid was in 
 great excess, which no doubt occasioned destructive 
 oxidation of some of the alkarsin with separation of 
 metallic arsenic and formation of water, acetone, and 
 acetic acid. 
 
 The yield of crude alkarsin from the distillation of 
 100 parts of arsenious acid and 100 parts of acetate 
 of potash is recorded as 30 parts, which indicates that 
 about one-half of the methyl in the acetate of potash 
 originally taken had survived so as to form the 
 ultimate arsenical organo-metallic liquid. 
 
 The product which distilled over is recorded as 
 consisting of an upper aqueous layer containing water, 
 acetone, and acetic acid, and a lower oleaginous layer 
 not miscible with the aqueous layer, and some solid 
 metallic arsenic at the bottom of all. The evolution 
 of abundance of carbonic acid during the distillation, 
 and the finding of the carbonate of potash in the residue 
 in the retort have likewise been duly recorded. 
 
 The important product is the heavy oily liquid 
 — the fuming liquor of Cadet, which Bunsen most 
 carefully washed, dried, and distilled in a current of 
 hydrogen gas, so as to avoid any oxidation from the 
 atmosphere, 
 
 This liquid was found to consist mainly of oxide of 
 
32 ARSENIC. 
 
 kakodyl, As 2 (CH 3 ) 4 0, but contained a little free 
 kakodyl, As 2 (CH 3 ) 4 , which rendered it spontaneously 
 inflammable in contact with the air. 
 
 Taking this liquid as his starting-point, Bunsen 
 carried out the celebrated research which disclosed the 
 existence of kakodyl and the compounds and deriva- 
 tives of kakodyl, opened up wide fields of undreamt 
 of chemistry, and imparted a degree of certainty and 
 precision to the initial stages of organic chemistry the 
 influence of which has been felt up to the present 
 time. 
 
 When kakodyl was discovered organic chemistry was 
 very young, and though chemists had visions of new 
 organic radicals (cyanogen, for instance, had been dis- 
 covered by Gay-Lussac in 18 1 5), the notion of atomicity 
 and organic types and series was undeveloped ; and 
 kakodyl and its compounds and derivatives wore a dress 
 somewhat different from that which they would have 
 assumed if they had been discovered within the last 
 few years. The following are some members of the 
 kakodyl series : — 
 
 Oxide of kakodyl . . As 2 (CH 3 ) 4 = As(CH 3 ) 2 As(CH 3 ) 2 
 
 Radical kakodyl . . As 2 (CH 3 ) 4 = As(CH 3 ) 2 As(CH 3 ) 2 
 
 Chloride of kakodyl . As(CH 3 ) 2 Cl 
 
 Trichloride of kakodyl As(CH 3 ) 2 Cl 3 
 
 Kakodylic acid . . . As(CH 3 ) 2 OOH 
 
 Kakodyl, As 2 C 4 H 12 = As(CH 3 ) 2 As(CH 3 ) 2 , the radical 
 of the series, which, in times gone by, was described 
 as being " the most perfect type of an organic quasi- 
 
CHEMISTRY OF ARSENIC. 33 
 
 metal which chemistry yet possesses," is obtained from 
 Cadet's fuming liquor by heating it with strong hydro- 
 chloric acid, which converts that liquid mainly into 
 chloride of kakodyl, and then decomposing the chloride, 
 which is an oily product, by means of metallic zinc, 
 which gives a solid compound of chloride of zinc 
 with kakodyl. Treatment of this solid compound with 
 water gives solution of chloride of zinc and free 
 kakodyl. 
 
 It is a colourless, transparent, thin liquid, heavier 
 than water, and non-miscible with and almost absolutely 
 insoluble in water. It dissolves readily in alcohol and 
 ether. 
 
 Its boiling-point is about 170 Centigrade, and it 
 solidifies at 6° Centigrade, crystallising in large trans- 
 parent square prisms. 
 
 It possesses a most peculiar and offensive smell, and 
 its vapours are exceedingly poisonous. It takes fire 
 spontaneously on exposure to the air, and should be 
 kept either in hermetically sealed vessels or else under 
 water carefully freed from air by previously boiling. 
 
 In contact with chlorine gas it takes fire, and then 
 there is irregular and destructive chemical action ; 
 but when the chlorine is dissolved in water, so as 
 to moderate the action, then the chemical change is 
 quite regular, thus : — 
 
 Kakodyl. Chloride of Kakodyl. 
 
 As(CHc 
 
 mch:):} +ci ^ As(CI%cl 
 
 c 
 
34 ARSENIC. 
 
 With bromine and iodine there are parallel reac- 
 tions. 
 
 Kakodyl dissolves sulphur and combines with it, 
 producing sulphuret of kakodyl. 
 
 When air is slowly admitted to kakodyl oxygen is 
 taken up by the kakodyl and the oxide formed thus : — 
 
 Kakodyl. Alkarsin. 
 
 2 
 
 and by the further action of the atmosphere kakodylic 
 acid is slowly produced. 
 
 Kakodyl, kept in sealed vessels, out of contact- with 
 the air, is a very stable body. It may be considerably 
 heated without decomposition, but, at about 400 Centi- 
 grade, it is resolved into metallic arsenic and a mixture 
 of marsh gas and olefiant gas, thus : — 
 
 As 2 (CH 3 ) 4 = As 2 + 2CH, + C 2 H 4 
 
 and there is no deposit of carbon. 
 
 Kakodyl has a vapour density of 7.1 01 as found by 
 Bunsen's experiment. The theory requires 7.27 for 
 the formula As 2 C 4 H 10 . Its percentage composition 
 is : — 
 
 Theory. 
 
 As., .... 
 C 4 . . . . 
 H 12 .... 
 
 2 IO 
 
 150 
 
 71-43 
 
 48 
 
 22.86 
 
 12 
 
 571 
 
CHEMTSTRY OF ARSENIC. 35 
 
 Although this substance contains as much as 71 per 
 cent, of arsenic, none of the ordinary tests for arsenic 
 reveal the presence of arsenic in it, except such tests 
 as involve raising it to temperatures bordering on low 
 red heat. Sulphuretted hydrogen does not give the 
 yellow sulphuret of arsenic, zinc and hydrochloric acid 
 do not give arseniuretted hydrogen. It is a typical 
 organo-metallic compound, the union of the arsenic 
 with the carbon or hydro-carbon being so complete that 
 the whole group AsC 2 H 4 behaves like an atom of 
 hydrogen, forming an organic quasi-metal. 
 
 Since the time of the kakodyl-research the chemistry 
 of organo-metallic bodies has been greatly extended, and 
 the arsenical examples have been greatly multiplied. 
 
 Cahours and Eiche have, for instance, obtained 
 arsentrimethyl, As(CH 3 ) 3 , a liquid boiling at 120 
 Centigrade ; and Baeyer has also added to the number, 
 and brought beautiful order into the chemistry, of these 
 arsenical organo-metallic substances. 
 
 Section 5. — Kakodylic Acid and its Salts. 
 
 Kakodylic acid, upon which great expectations are 
 founded, AsC 2 H 7 2 = As(CH 3 ) 2 OOH, is prepared by 
 oxidising kakodyl, or its oxide alkarsin, by means of 
 peroxide of mercury in contact with water. 
 
 The action is very energetic, aud care must be taken 
 to avoid too great heating of the materials. An excess 
 of peroxide of mercury is placed in water in a flask 
 
36 ARSENIC. 
 
 which is kept cool by being immersed in cold water. 
 Then the kakodyl or alkarsin is carefully added, being 
 run in under water. By suitable agitation, and by 
 the external application of cold water to the flask, 
 the chemical action is kept under proper control. The 
 first product is kakodylate of mercury and metallic 
 mercury, thus : — 
 
 j > 
 
 As 2 (CH 3 ) 4 + 4HgO = As. 2 (CH 3 ) 4 4 Hg + Hg 3 . 
 
 Then more kakodyl is carefully added to decompose 
 the kakodylate of mercury so long as metallic mercury 
 is caused to separate, and until a slight smell of 
 kakodyl becomes perceptible. The kakodylic acid dis- 
 solves in the water, and its solution may be evaporated 
 down and the residue dissolved in alcohol, from which 
 it crystallises beautifully in large colourless crystals. 
 It is not so soluble in alcohol as in water, and in 
 ether it is quite insoluble. 
 
 The yield of the acid is good. From 76 grammes 
 of Cadet's fuming liquor Bunsen obtained 88 grammes 
 of kakodylic acid. The theoretical yield from 76 
 grammes of oxide of kakodyl, As 2 (CH 3 ) 4 0, is 92.8 
 grammes of the acid. 
 
 The stability of kakodylic acid is wonderful. Nitric 
 acid does not attack it, nor yet sulphuric acid mixed 
 with chromate of potash. It is not reduced by sul- 
 phurous acid, nor by oxalic acid, nor by ferrous salts. 
 But phosphorous acid, acidulated stannous solution, or 
 zinc act upon it, and kakodyl is produced. 
 
CHEMISTRY OF ARSENIC. 37 
 
 Dry hydriodic acid reduces it thus : — 
 
 AsC 2 H 7 2 + 3HI = AsC 2 H 6 1 + 2H 2 + 1 2 . 
 
 Sulphuretted hydrogen, either in a dry condition or 
 dissolved in water, attacks kakodylic acid with vigour, 
 developing heat and producing bisulphuret of kakodyl 
 
 2 As0 2 H 7 2 + 3H 2 S = AsC 2 H 6 AsC 2 H 6 S 2 + S + 4.H 2 0. 
 
 If weak alcohol be taken instead of water, the 
 reaction is slightly different, thus : — 
 
 2 AsC 2 H 7 2 + 3H 2 S - AsC 2 H 6 AsC 2 H 6 S + S 2 + 4H 2 0. 
 
 Kakodylic acid has furnished the following on being 
 analysed : — 
 
 
 Calculated. 
 
 c, . 
 
 • 24 17.39 
 
 H r . 
 
 • 7 5-°7 
 
 As . 
 
 • 75 54-35 
 
 O s . 
 
 • 32 23.19 
 
 
 138 100.00 
 
 The salts of kakodylic acid are as a rule very 
 soluble in water. They seldom crystallise, and on 
 evaporating down their solutions in water they form 
 gummy masses. 
 
 According to Laurent the ammonia-salt does not 
 exist. The potash-salt is very deliquescent. The 
 soda-salt is less deliquescent, and has been made on a 
 considerable scale for medicinal use. 
 
38 ARSENIC, 
 
 Kakodylic acid forms crystalline compounds with 
 hydrofluoric and hydrochloric acids. The compound 
 with hydrochloric acid, AsC 2 H 7 2 HCl, is prepared by 
 dissolving kakodylic acid in strong hydrochloric acid 
 and leaving the solution in vacuo. It forms a mass of 
 crystals, and is without smell, but very acid to the 
 taste. Water resolves it into hydrochloric acid and 
 kakodylic acid. 
 
CHAPTER II. 
 
 DETECTION AND MEASUREMENT OF ARSENIC. 
 
 Section 1. — Marsh 1 s Test. 
 
 The Marsh's test, named after Marsh, who discovered 
 it, occupies a unique position in arsenical testing owing 
 to its surpassing delicacy and almost absolute appli- 
 cability. The qualification "almost" is called for by 
 reason of the existence and importance of the in- 
 stances wherein arsenic has entered into organo- 
 metallic relations with the organic elements and 
 rendered itself irresponsive to most of the general 
 tests for arsenic. 
 
 The justly celebrated James Marsh, chemist in the 
 Arsenal at Woolwich, and also chemical assistant at 
 the military school under Faraday, was born about the 
 year 1790, and died at Woolwich, 21st June 1846. 
 His discovery of the famous Marsh's test was brought 
 before the Society of Arts in London and published in 
 Jameson's Neiu Edinburgh Philosophical Journal, vol. 
 xxi., in the year 1836. 
 
 The author of this treatise — bearing in mind the 
 
 current ignorance, and the disparagement of this 
 
 process, and the misapprehension which is rife — re- 
 
 39 
 
4<D ARSENIC. 
 
 publishes Marsh's own words, quoting them from the 
 journal just mentioned. 
 
 "Account of a. Method of separating small quantities 
 of Arsenic from substances with which it may be 
 mixed, by James Marsh, Esq., of the Royal Arsenal, 
 Woolwich (communicated to the Society of Arts of 
 London). 
 
 "Notwithstanding the improved methods that have 
 of late been invented of detecting the presence of 
 small quantities of arsenic in the food, in the contents of 
 the stomach, and mixed with various other vegetable and 
 animal matters, a process was still wanting for separat- 
 ing it expeditiously and commodiously, and presentiug 
 it in a pure unequivocal form for examination by 
 the appropriate tests. Such a process should be 
 capable of detecting arsenic not only in its usual state 
 of white arsenic or arsenious acid, but likewise in that 
 of arsenic acid and of all the compound salts formed 
 by. the union of either of these acids with alkaline 
 substances. 
 
 "It ought also to exhibit the arsenic in its reguline 
 or metallic state free from the ambiguity which is 
 sometimes caused by the use of carbonaceous reducing 
 fluxes. It appeared to me that these objects might 
 be attained by presenting to the arsenic hydrogen 
 gas in its nascent state, the first action of which 
 would be to deoxygenate the arsenic; and the next 
 to combine with the arsenic thus deoxygenated into 
 
DETECTION AND MEASUREMENT OF ARSENIC. 4 1 
 
 the well-known gas called arseniu retted hydrogen. 
 Being thus brought to the gaseous state the 
 arsenic would spontaneously (so to speak) separate 
 itself from the liquor in which it was before dis- 
 solved, and might be collected for examination by 
 means of any common gas apparatus, thus avoiding 
 the trouble, difficulty, and ambiguity of clarification 
 and other processes whereby liquors suspected of con- 
 taining arsenic are prepared for the exhibition of the 
 usual tests, or of evaporation and deflagration which 
 are sometimes had recourse to in order to separate 
 the arsenic from the organic substances with which 
 it may have been mixed. I had the satisfaction of 
 finding on trial that my anticipations were realised, 
 and that I was thus able not only to separate very 
 minute quantities of arsenic from gruel, soup, porter, 
 coffee, and other alimentary liquors, but that by con- 
 tinuing the process a sufficient length of time I could 
 eliminate the whole of the arsenic in the state of 
 arseniuretted hydrogen, either pure or at most only 
 mixed with an excess of hydrogen. 
 
 " If this gas be set fire to as it issues from the end of 
 a jet of fine bore into the common air the hydrogen, as 
 the more combustible ingredient, will burn first, and 
 will produce aqueous vapour, while the arsenic will be 
 deposited either in the metallic state or in that of 
 arsenious acid, according as it is exposed partially or 
 freely to the air. The former condition is brought 
 about by holding a piece of cold window-glass opposite 
 
42 ARSENIC. 
 
 to and in contact with the flame, when a thin metallic 
 film will be immediately deposited on its surface ; and 
 the latter by receiving the flame within a glass tube 
 open at both ends., which in half a minute will be found 
 to be dimmed by a white pulverulent sublimate of 
 arsenious acid. By directing the flame obliquely upon 
 the inside of the tube it strikes against the glass and 
 deposits the arsenic partly in the metallic state. In 
 this case, if the tube while still warm be held to the 
 nose, that peculiar odour somewhat resembling garlic 
 which is one of the characteristic tests of arsenic will 
 be perceived. Arseniuretted hydrogen itself has pre- 
 cisely the same colour, but considerable caution should 
 be used in smelling it, as every cubic inch contains 
 about a quarter of a grain of arsenic. 
 
 The requisite apparatus is as simple as possible, 
 being — 
 
 " A glass tube open at both ends and about three- 
 quarters of an inch in its internal diameter. It is bent 
 in the form of a syphon (« a), the shorter leg being 
 about 5 inches and the longer leg about 8 inches in 
 length. A stopcock (b) ending in a jet of fine bore 
 passes tightly through a hole made in the axis of a soft 
 and sound cork, which fits airtight into the opening of 
 the lower bend of the tube, may be further secured, if 
 requisite, by a little common lute. To fix the apparatus, 
 when in use. in the upright position a hole is made in 
 the wooden block c for the reception of the pillar d. 
 and a groove is cut in the top of the same block to 
 
DETECTION AND MEASUREMENT OF ARSENIC. 43 
 
 receive the bend of the tube a a. Two elastic, slips 
 (e e) cut from the neck of a common bottle of india- 
 rubber keep the tube firm in its place. 
 
 " The matter to be submitted to examination and 
 supposed to contain arsenic, if not in the fluid state, such 
 as pastry, pudding, or bread, &c, must be boiled with 
 two or three fluid ounces of clean water for a sufficient 
 length of time. The mixture so obtained must then be 
 thrown in a filter to separate the more solid parts ; 
 thick soup or the contents of the stomach may be 
 diluted with water and also filtered ; but water, gruel, 
 wine, spirits, or any kind of malt liquor, and such like 
 as tea, coffee, cocoa, &c, can be operated on without 
 any previous process. 
 
 M When the apparatus is to be used a bit of glass rod 
 about an inch long is to be dropped into the shorter leg, 
 and this is to be followed by a piece of clean sheet-zinc 
 about an inch and a half loug and half an inch wide, 
 bent double so that it will run down the tube till it is 
 stopped by the piece of glass rod first put in. The 
 stopcock and jet are now to be inserted and the handle 
 is to be turned so as to leave the cock open. The 
 fluid to be examined having been previously mixed 
 with from a drachm and a half to three drachms of 
 dilute sulphuric acid (1 acid and J water), is to be 
 poured into the long leg till it stands in the short one 
 about a quarter of an inch below the bottom of the 
 cork. Bubbles of gas will soon be seen to rise from the 
 zinc which are pure hydrogen if no arsenic be present; 
 
44 ARSENIC. 
 
 but if the liquor holds arsenic in any form in solution 
 the gas will be arseniuretted hydrogen. The first 
 portions are to be allowed to escape in order that they 
 may carry with them the small quantity of common air 
 left in the apparatus ; after which the cock is to be 
 closed, and the gas will be found to accumulate in the 
 shorter leg, driving the fluid up the longer one till the 
 liquor has descended in the short leg below the piece 
 of zinc, when all further production of gas will cease. 
 There is thus obtained a portion of gas subject to the 
 pressure of a column of fluid of from seven to eight 
 inches high ; when therefore the stopcock is opened the 
 gas will be propelled with some force through the jet, 
 and on igniting it as it issues (which must be done 
 quickly by an assistant), and then holding horizontally 
 a piece of crown or window glass over it in such a 
 manner as to retard slightly the combustion, the arsenic 
 (if any be present) will be found deposited in the 
 metallic state on the glass, the oxygen of the atmos- 
 phere being employed in oxidising the hydrogen only 
 during the process. If no arseuic be present then the 
 jet of the flame as it issues has a very different appear- 
 ance ; and, although the glass becomes dulled in the 
 first instance by the deposition of the newly formed 
 water, yet such is the heat produced that in a few 
 seconds it becomes perfectly clear aud frequently flies 
 to pieces. 
 
 "If the object be to obtain the arsenic in form of 
 arsenious acid or white arsenic then a glass tube from 
 
DETECTION AND MEASUREMENT OF ARSENIC. 45 
 
 a quarter to half an inch in diameter (or according to 
 the size of the jet of flame) and eight or ten inches in 
 length is to be held vertically over the burning jet of 
 gas in such a manner that the gas may undergo perfect 
 combustion, and that the arsenic combined with it may 
 become sufficiently oxidised ; the tube will thus with 
 proper care become lined with arsenious acid in pro- 
 portion to the quantity originally contained in the 
 mixture. 
 
 " When the glass tube is held at an angle of about 
 45 over the jet of flame three very good indications of 
 the presence of arsenic may be obtained at one opera- 
 tion, viz. metallic arsenic will be found deposited 
 in the tube at the part nearest where the flame im- 
 pinges — white arsenic or arsenious acid at a short 
 distance from it — and the garlic smell can be readily 
 detected at either end of the tube in which the 
 experiment has been made. 
 
 "As the gas produced during the operation is con- 
 sumed the acid mixture falls into the short limb of 
 the tube, and is thus again brought into contact with 
 the zinc, in consequence of which a fresh supply is 
 soon obtained. This gas, if submitted to either of the 
 processes before described, will give fresh indications 
 of the presence of the arsenic which the mixture may 
 have originally contained, and it will be easily per- 
 ceived that the process may be repeated as often as 
 may be required at the will of the operator till no 
 further proofs can be obtained. 
 
46 AKSENIC. 
 
 "When certain mixed or compound liquors are 
 operated on in the apparatus a great quantity of 
 froth is thrown up into the tube, which may cause 
 a little embarrassment by choking the jet. I have 
 found this effect to take place most with the contents 
 of the stomach, with wine, porter, tea, coffee, or soup, 
 and indeed with all mucilaginous and albuminous 
 mixtures. The means I adopt to prevent this effect 
 from taking place, or at least for checking it in a 
 great measure, is to grease or oil the interior of the 
 short limb of the apparatus before introducing the 
 substance to be examined, or to put a few drops of 
 alcohol or sweet oil on its surface previously to intro- 
 ducing the stopcock and its appendages. I have 
 however found, if the tube be ever so full of froth 
 in the first instance, that in an hour or two if 
 left to itself the bubbles burst, and the interior of 
 the tube becomes clear without at all affecting the 
 results. 
 
 " In cases where only a small quantity of the matter 
 to be examined can be obtained, I have found a great 
 convenience in using the small glass bucket g. Under 
 such circumstances the bent glass tube may be filled 
 up to within an inch of the short end with common 
 water, so as to allow room for the glass bucket, which 
 must be attached to the cork c by means of a little 
 platina wire. A bit or two of zinc is to be dropped 
 into the bucket with a small portion of the matter to 
 be examined, and three or four drops of diluted sul- 
 
DETECTION AND MEASUREMENT OF ARSENIC. 47 
 
 phuric acid (acid 2, water 14), and the whole is then 
 to be introduced into the mouth of the short limb of 
 the tube. The production of gas under this arrange- 
 ment is much slower, and of course requires more time 
 to fill the tube than in the former case, but the mode 
 of operating is precisely the same. Indeed it is of 
 great advantage, when the quantity of arsenic present 
 is very minute, not to allow the hydrogen to be evolved 
 too quickly, in order to give it time to take up the 
 arsenic. A slender glass funnel will be found of 
 service when as much as a table-spoonful or even a 
 teaspoonful of matter can be obtained for examination. 
 In this case the tube is to be partly filled with common 
 water, leaving a sufficient space for the substance to 
 be examined ; a piece of zinc is to be suspended from 
 the cork by a thread or wire so as to hang in the axis 
 of the tube, and the fluid to be operated on having 
 been previously mixed with dilute sulphuric acid, is 
 then to be poured through the funnel carefully, so as 
 to surround the zinc, avoiding as far as possible to mix 
 it with the water below, and the stopcock and its 
 appendages are to be replaced in the mouth of the 
 tube. The production of the gas then goes on as 
 before stated, and the mode of manipulating with it 
 is exactly the same as described in the foregoing part 
 of the paper. It will be necessary for me in this place 
 to explain the methods I employ after each operation 
 to determine the integrity of the instrument, so as to 
 satisfy myself that no arsenic remains adhering to the 
 
48 AKSENIC. 
 
 inside of the tube or to the cork and its appeudages 
 before I employ it in another operation. After washing 
 the apparatus with clean water a piece of zinc may be 
 dropped in, and the tube filled to within half an inch 
 of the top of the short limb ; two drachms of diluted 
 sulphuric acid are then poured in, and the stopcock 
 and cork secured in its place. Hydrogen gas will in 
 this case, as before, be liberated and fill the tube. If 
 the gas as it issues from the jet be then inflamed and 
 a piece of window-glass held over it as before described, 
 and any arsenic remaius, it will be rendered evident 
 by being deposited on that glass ; if so, this opera- 
 tion must be repeated till the glass remains perfectly 
 clean after having been exposed to the action of 
 the gas. 
 
 "When I have had an opportunity of working with 
 so large a quantity of mixture as from two to four 
 pints (imperial measure) I then have employed the 
 instrument here figured, which is indeed but a 
 slight modification of one of the instantaneous light 
 apparatuses now well known and used for obtaining 
 fire by the aid of a stream of hydrogen gas thrown on 
 spongy platinum. It will therefore be of importance 
 only for me to describe the alteration which I make 
 when I employ it for the purpose of detecting arsenic. 
 In the first place, I must observe that the outer vessel 
 (a) which I use holds full four pints, and that the jet 
 of the stopcock is vertical, and its orifice is twice or 
 three times larger than in the instrument as generally 
 
DETECTION AND MEASUREMENT OF ARSENIC. 49 
 
 made for sale, and also that there is a thread or wire 
 attached to the cork of the stopcock (b) for suspending 
 a piece of zinc (c) within the bell glass. 
 
 "With an instrument of this description I have 
 operated on 1 grain of arsenic in 28,000 grains of 
 water (or four imperial pints), and have obtained 
 therefrom upwards of 100 distinct metallic arsenical 
 crusts. Similar results have been obtained with perfect 
 success from three pints of very thick soup, the same 
 quantity of port wine, porter, gruel, tea, coffee, &c. &c. 
 
 "It must be understood that the process was slow 
 and required several days before the mixture used 
 ceased to give indication of the presence of arsenic, 
 and also a much larger portion of zinc and sulphuric 
 acid was employed from time to time than when work- 
 ing with the small bent tube apparatus, in consequence 
 of the large quantity of matter operated on under this 
 arrangement. 
 
 " With the small apparatus I have obtained distinct 
 metallic crusts when operating on so small a quantity 
 as one drop of Fowler's solution of arsenic, which only 
 contains T |-o tn P arfc °^ a grain. The presence of 
 arsenic in artificial orpiment, and realgar, in Scheele's 
 Green and in the sulphuret of antimony, may be 
 readily shown by this process when not more than 
 half a grain of any of those compounds is em- 
 ployed. In conclusion, I beg to remark that although 
 the instruments I have now finished describing are the 
 form I prefer to all I have employed, it must be 
 
50 ARSENIC. 
 
 perfectly evident to any one that many very simple 
 arrangements might be contrived. Indeed I may say 
 unequivocally there is no town or village in which 
 sulphuric acid and zinc can be obtained but every 
 house could furnish to the ingenious experimental- 
 ist ample means for the purpose. For a two-ounce 
 phial, with a cork and a piece of tobacco-pipe, or 
 a bladder with the same arrangement fixed to its 
 mouth might, in cases of extreme necessity, be 
 employed with success, as I have repeatedly done 
 in this instance. The only ambiguity that can pos- 
 sibly arise in the mode of operating above de- 
 scribed, arises from the circumstance that some 
 samples of zinc of commerce themselves contain 
 arsenic ; and such when acted on by dilute sulphuric 
 acid gave out arseniuretted hydrogen. It is therefore 
 necessary for the operator to be certain of the purity 
 of the zinc which he employs, and this is easily done 
 by putting a bit of it into the apparatus with only 
 some dilute sulphuric acid, the gas thus obtained to 
 be set fire to as it issues from the jet, and if no 
 metallic film is deposited on the bit of flat glass, 
 and no white sublimate within the open tube, the 
 zinc may be regarded as in a fit state for use." 
 
 Doubtless some of the readers of this book will be 
 surprised to find that Marsh's process, when it left 
 the hands of the master, was capable of dealing with 
 mixtures of arsenious acid with the various common 
 articles of food and drink without demanding, as a 
 
DETECTION AND MEASUREMENT OF ARSENIC. 5 1 
 
 preliminary operation, the destruction of the organic 
 matter. And the author of this book expects that 
 most chemists will be surprised to learn that Marsh 
 himself claimed for his method that it " eliminates 
 the whole of the arsenic in the state of arseniuretted 
 hydrogen." In order that the elimination may be 
 complete, it is essential to conduct the operation in 
 a proper manner ; and a careful and intelligent study 
 of the text of Marsh's paper (which has just been 
 quoted without alteration) will enable the reader to 
 verify the justice of the claims which have been 
 brought forward. It will be noted that one of the 
 conditions of success is that the development of 
 hydrogen must be very gradual and protracted. 
 
 In France and Germany Marsh's test has been well 
 appreciated and likewise developed in a remarkable 
 manner. When the stream of hydrogen (laden as it is 
 with arseniuretted hydrogen) issues from the Marsh 
 apparatus, there are other ways of utilising it besides 
 burning it or passing it through a red-hot tube. The 
 gas may be passed into various solutions containing salts 
 of the various heavy metals, and most interesting and 
 important are the chemical events which come to pass. 
 
 Thus, when in the year 1 874 in France, Mayengon 
 and Bergeret passed the stream of gas, as it issued 
 from the Marsh apparatus, into a solution of corrosive 
 sublimate, they reported that the following change 
 took place : — 
 
 6HgCl 2 +2AsH 3 =3Hg 2 Cl 2 + As 2 + 6HCl. 
 
52 ARSENIC. 
 
 Solution of perchloride of mercury (corrosive sub- 
 limate) and arseniuretted hydrogen are resolved into 
 subchloride of mercury (calomel) and metallic arsenic 
 and hydrochloric acid. The action is prompt, and is 
 available for rendering quantitative results. 
 
 The mixed calomel and metallic arsenic form a 
 precipitate, which by filtration may be collected upon 
 filter paper, and subsequently weighed and investigated 
 in many ways which would suggest themselves to a 
 chemist. 
 
 From an analytical point of view this manner of 
 dealing with arseniuretted hydrogen offers a very 
 striking advantage. Make the calculation by replacing 
 the chemical symbols in the equation by the proper 
 numbers, and it will be found that 150 parts of 
 metallic arsenic received in the state of arseniuretted 
 hydrogen by the solution of corrosive sublimate yields 
 1563 parts of mixed calomel and metallic arsenic. 
 
 If arsenic were estimated in that way, one unit of 
 arsenic would be represented by 10.42 of the solid to 
 be weighed. 
 
 An investigation made in the author's laboratory 
 has confirmed the substantial accuracy of the report. 
 The unit of arsenic in the shape of arsenious acid is 
 represented by rather more than 10.42 units of the 
 dry precipitate. The author is of opinion that the 
 metallic arsenic is not in mixture with the calomel, 
 but in real chemical combination. 
 
 The two French chemists above mentioned have 
 
DETECTION AND MEASUREMENT OF ARSENIC. 53 
 
 employed this method as a means of distinguishing 
 between the antimony and the arsenic compounds with 
 hydrogen, both of which are gases. 
 
 With that object in view they place a little of the 
 solution of corrosive sublimate on glazed white paper, 
 and hold the paper in the stream of gas. If arseni- 
 uretted hydrogen be there, a citron-yellow stain makes 
 its appearance, and after a while the stain assumes a 
 brown-yellow colour. Antimoniuretted hydrogen, on 
 the other hand, strikes a dark grey colour which 
 cannot be mistaken for the arsenical spot. Experi- 
 ments on the delicacy of the method have been pub- 
 lished showing that solutions with one part of arsenious 
 acid in 70,000 parts of the solution strikes the citron- 
 yellow colour in one minute, and that solutions with 
 one part in 120,000 require five minutes in order to 
 produce the stain. 
 
 Arseniuretted hydrogen and solution of nitrate of 
 silver react as follows : — 
 
 6AgN0 3 + AsH 3 + 3 H 2 = 6HN0 3 + Ag 6 + As0 3 H 3 . 
 
 The As0 3 H 3 is, of course, resolved into water and 
 arsenious acid, thus : — 
 
 2(As0 3 H 3 ) = As 2 3 +3H 2 0. 
 
 When the gas issuing from Marsh's apparatus is 
 passed into the solution of nitrate of silver, the first 
 visible effect is that the silver-solution darkens, assum- 
 ing a brownish-black colour ; then, as the passage of 
 
54 ARSENIC. 
 
 the gas is continued, a voluminous black precipitate 
 separates from the liquid which gradually settles 
 down and contracts and consists of metallic silver. 
 The solution gradually becomes acid, and is found to 
 contain arsenious acid, the presence of which may be 
 manifested by adding with great caution, drop by 
 drop, dilute solution of ammonia, so as to neutralise 
 the liquid, whereupon the yellow arsenite of silver will 
 make its appearance. 
 
 The behaviour of the gas towards silver-solution 
 affords a means of distinguishing between the gaseous 
 compounds of arsenic and antimony. Arseniuretted 
 hydrogen throws down pure silver, and charges the 
 solution with arsenious acid : antimoniuretted hydrogen 
 throws down an alloy of antimony and silver, and no 
 antimonial compound enters into the solution. 
 
 The author of this book has carried out experiments 
 with the object of ascertaining the practicability or 
 otherwise of making estimations of arsenic by the 
 employment of this reaction in order to deal with the 
 stream of gas from Marsh's apparatus. On inspection 
 of the equation given above it is manifest that every 
 atom of arsenic passed into the nitrate of silver should 
 reduce six atoms of silver ; therefore, every unit of 
 arsenic is represented by 9. 1 2 units of metallic silver. 
 As a product of the reaction there is the arsenious 
 acid in the solution, and that may be converted into 
 sulphuret of arsenic and weighed in that form, or else 
 it might be oxidised into arsenic acid and weighed as 
 
DETECTION AND MEASUREMENT OF ARSENIC. 55 
 
 arseniate of magnesia. The author has weighed the 
 precipitated silver, filtering off the solution containing 
 the As 2 3 , oxidised and finally weighed the arsenic 
 acid as salt of magnesia. 
 
 Many difficulties of detail stood in the way, but they 
 have been so far overcome that the author has no 
 hesitation in making the announcement that the Marsh 
 test, worked by passing the gas through solution of 
 nitrate of silver, affords a useful quantitative determina- 
 tion of the arsenic. 
 
 The initial difficulty was, of course, how to prevent 
 the greater part of the arsenic from becoming masked by 
 passing into states of existence other than the state of 
 the gaseous compound of arsenic and hydrogen. One of 
 the devices resorted to in order to avoid that calamity 
 was to keep the acid in the Marsh apparatus in slight 
 excess over the zinc. Strict account is kept of the 
 quantities of zinc and acid employed ; and the form of 
 the apparatus has been somewhat altered in order to 
 facilitate the different steps which have to be taken. 
 
 Another point requiring attention was how to insure 
 complete action between the gaseous arsenic compound 
 and the silver-solution. The gas proceeding from a 
 Marsh's apparatus is, of course, dilute arseniuretted 
 hydrogen — very dilute indeed in most cases. Unless 
 proper precautions be taken the vast volume of free 
 hydrogen will protect the arseniuretted hydrogen from 
 the silver-solution, and in that way the determination 
 of arsenic would be damaged. 
 
56 ARSENIC. 
 
 The strength of the silver-solution has also received 
 attention, and a one per cent, solution, i.e. one gramme 
 of nitrate of silver in ioo c. c. of distilled water, has 
 been found to answer. 
 
 The internal capacity of the gas-generating vessel 
 employed in these experiments was 270 c. c. : the 
 generator being a very strong glass bottle with a 
 strong neck into which fitted a good cork properly 
 bored. The granulated zinc employed was carefully 
 weighed and gradually introduced ; and some idea of 
 the ratio of hydrogen to arsenic volatilised may be 
 gathered from the statement that 14.56 grammes of 
 zinc were dissolved in the generating vessel and O.494 
 gramme of reduced silver was weighed. After that 
 had taken place a further introduction of 2.34 grammes 
 of zinc with the necessary acid occasioned no further 
 ponderable reduction of silver. 
 
 The study of the chemical action which takes place 
 in Marsh's apparatus leads to the conclusion that the 
 reduction of As 2 3 so as to yield up the arsenic in the 
 form of 2AsH 3 proceeds by stages. 
 
 The equation 
 
 As 2 3 + 6H 2 = 2(AsH 3 ) + 3H 2 
 expresses the final effect ; but there are evidently 
 stages in the operation carried on in the gas-generating 
 vessel. A first portion of the gas comes over with 
 great facility, and direct experiment has proved that 
 much, if not absolutely all of the masked arsenic, 
 remains in the clear solution. 
 
DETECTION AND MEASUREMENT OF ARSENIC. $7 
 
 There is no difficulty in converting half of the 
 arsenic existing in arsenious acid into arseniuretted 
 hydrogen, but to obtain nearly the whole in the form 
 of arseniuretted hydrogen (though quite possible) is 
 rather tedious. 
 
 The author offers the following suggestion. First 
 stage of the reaction : — 
 
 As 2 3 + 4 H 2 = AsH 3 + AsH 2 0. 2 H + H 2 0. 
 
 The hypothetical body AsH 2 2 H, would be the 
 hydrogen representative of kakodylic acid, slowly 
 attacked by long-continued hydrogenation in the 
 Marsh's apparatus. Whether this hypothesis be the 
 true theory of the changes or not, the fact is established 
 that whether the scale be great or small there is a first 
 outburst of arseniuretted hydrogen until about half of 
 the possible arseniuretted hydrogen has been given off, 
 and then the yield slows down and ultimately stops. 
 
 The theory derives some support from an experi- 
 ment in which to begin with the arsenic must have 
 been in the state of arsenic acid, and when for a 
 while there was almost absolutely no evolution of 
 arseniuretted hydrogen and then a slow evolution of 
 the arsenical gas. 
 
 The author is studying the action of different solu- 
 tions on the gas from the Marsh apparatus. So far as 
 his experiments have gone, he prefers the solution of 
 corrosive sublimate. 
 
58 ARSENIC. 
 
 Section 2. — ReinscKs Test. 
 
 Edgar Hugo Emil Keinsch, born in the year 1809, 
 was the discoverer of the well-known Eeinsch's test, 
 apparently about the year 1842. He died in Erlangen, 
 1st November 1884. 
 
 In Fownes's " Manual of Elementary Chemistry," 
 published in the year 1844, there is the following 
 passage on page 316: — 
 
 "A slip of copper-foil boiled in the poisoned liquid, 
 previously acidulated with hydrochloric acid, withdraws 
 the arsenic and becomes covered with a white alloy. 
 By heating the metal in a glass tube the arsenic is 
 expelled." 
 
 At that date Fownes was Chemical Lecturer in the 
 Middlesex Hospital Medical School and to the Pharma- 
 ceutical Society of Great Britain, and his excellent and 
 most interesting manual was, no doubt, the means of 
 introducing this test to Englishmen. By an oversight, 
 no doubt, purely accidental, the name of Reinsch was 
 not mentioned by Fownes, the passage above quoted 
 being the concluding paragraph of the section on 
 arsenic, and immediately following the detailed account 
 of the Marsh's test, which was duly ascribed to Mr. 
 Marsh. 
 
 The seeming simplicity of this test at once com- 
 mended it to attention in England, and Taylor of 
 Guy's Hospital made abundant use of it in medico- 
 legal inquiries. 
 
DETECTION AND MEASUREMENT OF AKSENIC. 59 
 
 The principle of the test is quite intelligible. 
 Arsenious acid As 2 3 , hydrochloric acid HOI, in some- 
 what dilute aqueous solution, and metallic copper Cu 2 , 
 finding themselves together, react thus : — 
 
 3 Cu, + As 2 3 + 6HC1 = 3 H 2 + 6C11CI + As 2 . 
 
 The 6C11CI dissolves in the hydrochloric acid and the 
 As 2 combines with the surface of the copper-foil, pro- 
 ducing the white-grey alloy of copper and arsenic, and 
 clings to the copper. On heating the coated copper- 
 foil to low redness the alloy is decomposed and the 
 arsenic sublimes and oxidises, and may be collected on 
 the inside surface of a narrow glass tube in the form of 
 glittering crystal of As 2 3 . The copper-foil acts as a 
 sort of sponge for the collection of the arsenic which 
 exists in the liquid submitted to the operation of the 
 Keinsch's test. 
 
 It is quite true that for every equivalent of oxygen 
 in the arsenious acid an equivalent of copper passes 
 into solution, and to that extent the copper-foil is used 
 up, but for the rest the action of the copper is that of 
 a carrier, an arsenic-carrier. 
 
 In the year 1861, a distinguished German chemist of 
 the name of Werther published an important observa- 
 tion which he had made in working Eeinsch's test. He 
 found that solutions containing even a considerable 
 proportion of arsenic acid, or of salts of arsenic acid, 
 dissolved in hydrochloric or sulphuric acid did not at 
 all readily cause the well-known arsenical coating of the 
 
60 ARSENIC. 
 
 slip of copper-foil. Months of contact in the cold pro- 
 duced no arsenical coating ; alternate boiling and cool- 
 ing of the solutions with the slip of copper-foil immersed 
 in them were equally without effect. And the conclu- 
 sion arrived at by Werther was that only arsenious 
 acid and its salts (and not arsenic acid and its salts) 
 is capable of giving the arsenical coating to a slip of 
 bright copper-foil by short immersion in the boiling- 
 acid liquid. 
 
 Many quantitative experiments were recorded in 
 which time and proportions of arsenic were noted, and 
 the general result was that as much as one per cent, of 
 arsenic (700 grains per gallon) in the shape of arsenic 
 acid was needed in order that the slip of copper might 
 acquire an unmistakable arsenical coating by short 
 boiling in the acid liquid. 
 
 Werther showed also that a solution of cv/pric chloride 
 in hydrochloric acid dissolved metallic arsenic when the 
 solution was gently warmed. 
 
 Eeinsch himself has verified and confirmed much of 
 Werther's work. 
 
 Reinsch admits that when the dilute solution of 
 arsenic acid is mixed with half its volume, or even its 
 own volume of strong hydrochloric acid, the Reinsch 
 test utterly fails to detect traces of arsenic. 
 
 But Reinsch insists upon the validity of hi3 test 
 when the proportion of hydrochloric acid is so much 
 increased that there are two volumes of strong hydro- 
 chloric acid to one volume of the arsenical solution. 
 
DETECTION AND MEASUREMENT OF ARSENIC. 6 1 
 
 " The acid solution must be so strong that the copper 
 is attacked, otherwise the arsenic acid is not reduced." 
 Reinsch maintains that with a sufficiency of hydro- 
 chloric acid, i.e. two volumes of the strong acid to one 
 volume of the dilute solution of arsenic acid, the pro- 
 duction of the arsenical coating is almost instantaneous, 
 and that one part of arsenic is distinctly to be recog- 
 nised in 100,000 parts of the arsenical solution, just as 
 if it were present in the state of arsenious acid. In 
 the same year, viz. 1861, E. C. Stanford, in this country, 
 called attention to the fact that the presence of so 
 common a substance as perchloride of iron in the sup- 
 posed arsenical solution would bring about discolora- 
 tion of the slip of copper in the Reinsch's test. 
 Attention was likewise directed to the fact that most, 
 if not all, commercial copper is more or less arsenical, 
 and in 1863 a vei 7 curious paper was brought before 
 the Chemical Society of London by Dr. William 
 Odling, F.R.S. The paper bore the title "Note on 
 the Detection of Arsenic in Copper," and opened as 
 follows : — 
 
 " As even in the most satisfactory performance of 
 Reinsch's test for arsenic — the deservedly favourite 
 test of English toxicologists — there is always some, 
 although but an extremely small quantity of the 
 copper wire, foil, or gauze dissolved, and as com- 
 mercial copper is rarely quite free from arsenic, and 
 sometimes contains a very notable proportion thereof, 
 it is important that the copper to be used in medico- 
 
62 ARSENIC. 
 
 legal researches as a precipitant for arsenic should be 
 specially tested as to its purity." 
 
 After some further remarks Dr. Odling proceeds: — 
 "A few grains of the copper cut into small pieces 
 are placed in a small tube-retort with an excess of 
 hydrochloric acid and so much ferric hydrate or 
 chloride as contains a quantity of iron about double 
 the weight of the copper to be acted upon. The 
 mixture is then distilled to dryness, some care being 
 taken at the last to prevent spurting. The whole 
 of the copper is in this way quickly dissolved, and 
 any arsenic originally contained in it carried over in 
 the form of chloride of arsenic, which may be con- 
 densed in a little water with the excess of aqueous 
 hydrochloric acid. The resulting distillate is then 
 tested for arsenic." 
 
 In the next year, viz. 1864, Hellwig published in 
 Germany an account of the employment of the micro- 
 scope in toxicology, in which he described a mode of 
 procedure by which very minute quantities of arsenic, 
 deposited on glass or porcelain, might be sublimed 
 and in that way transferred to the slide of the micro- 
 scope, where the characteristic form of the crystals 
 of arsenious acid might be verified. Of course this 
 method is of general application whether the trace 
 of arsenical deposit were derived from the deposit 
 by Marsh's test or Eeinsch's test, or deposit pro- 
 duced in other ways. Hellwig especially noted that, 
 whereas an arsenical deposit would yield a sublimate 
 
DETECTION AND MEASUREMENT OF ARSENIC. 6$ 
 
 of As 2 3 crystals, a deposit of pure antimony gave 
 nothing. 
 
 In comparing Marsh's aud Reinsch's tests, the first 
 point is that neither the one nor the other is capable of 
 dealing with the case where the arsenic has entered 
 into the organo-metallic state. Applied to kakodylic 
 acid, for instance, both would fail. 
 
 But arsenious acid in mere mixture with organic 
 matter is to be recognised easily by either Marsh or 
 Reinsch. 
 
 Arsenic acid, under like conditions, is to be recog- 
 nised by Marsh's test, but not by Reinsch's test, as it 
 has been recently described in a very modern docu- 
 ment, to which reference will presently be made. By 
 Reinsch's test, when abundance of hydrochloric acid is 
 used, as Reinsch himself recommended in the year 
 1 36 1, arsenic acid may be recognised. The seeming 
 simplicity of the Reinsch test is more apparent than 
 real. The necessity of having the copper free from 
 arsenic is a source of trouble to the operator ; and 
 the advice may be given that before resorting to the 
 Reinsch test a preliminary experiment should be 
 made in order to see whether or not the copper-foil 
 itself is capable of throwing off arsenic and causing a 
 sublimate of arsenious-acid crystals on the microscope- 
 slide. 
 
 The hopelessness of the Reinsch test, from a quanti- 
 tative point of view, is in great part due to the cir- 
 cumstance that there can be no action of the test 
 
64 ARSENIC. 
 
 without some of the copper passing into solution ; and 
 the entrance of copper into solution arrests the deposit 
 of arsenic on the copper, and even, under some con- 
 ditions, brings about re-solution of arsenic already- 
 deposited. Werther's important investigation, an 
 abstract of which will be found in the Jahrcshericht 
 for the year 1861, to which reference has been made, 
 is very much to the point. The absolute absence of 
 any relation between the amount of discoloration of 
 the copper-foil and the quantity of arsenic in the 
 liquid is fatal. The employment of one very obvious 
 way of attempting a rough quantitative employment 
 of the test, coupled with the resort to the microscope 
 for the appreciation of the crystals of As 2 O s , brings in 
 its own peculiar difficulties and uncertainties. 
 
 The following recent experiment performed in the 
 laboratory of the author may be cited. A mixture of 
 200 c. c. of London water from the Lambeth Water 
 Company, 15 c. c. of a solution containing one milli- 
 gramme of As 2 3 per cubic centimetre of the solution, 
 and 30 c. c. strong hydrochloric acid was subjected to 
 the Eeinsch's operation in the manner approved by 
 the brewers' experts to which reference will presently 
 be made. First one lot of copper was introduced, and 
 in due time, after the prescribed boiling in contact 
 with the solution, it was taken out, washed, dried, and 
 heated, and abundant As 2 3 sublimed. Then a second 
 lot of copper was put into the solution and treated as 
 before, and As 2 3 obtained. And finally the residual 
 
DETECTION AND MEASUREMENT OF ARSENIC. 65 
 
 solution was placed in the Marsh apparatus, and still it 
 evolved abundance of arsenic. 
 
 Since the year 1864 some branches of chemical 
 analysis have decayed in England. The following 
 passage, quoted from the report of the commission of 
 experts appointed by the Manchester Brewers' Central 
 Association to investigate the circumstances of the 
 arsenical beer poisoning, affords a curious example 
 of such decay : — 
 
 "The most important matter for the moment is to 
 secure adequate testing of beer in order that the public 
 may be protected from further mishap. Arsenic is 
 a substance which can be detected in the most in- 
 finitesimal quantities by those who are practised in 
 the tests, but these tests are so delicate that they are 
 apt to mislead those who have not had experience in 
 their application, and this is more particularly the 
 case when the test is to be applied to a complex 
 mixture such as beer. Accordingly, the commission 
 have thought it necessary to investigate and determine 
 what is the most suitable method of testing beers for 
 arsenic. 
 
 " Test. — The commission recommend that the Reinsch 
 test should be used in preference to all others at pre- 
 sent known, because their investigations have satisfied 
 them that it is the best and most reliable test for 
 arsenic in beer. The mode of performing it is as 
 follows : Take 200 c. c. of the beer in a porcelain 
 evaporating-dish. Raise the liquid to the boiling-point, 
 
65 ARSENIC. 
 
 and then add 30 c. c. of pure concentrated hydrochloric 
 acid. Insert a piece of pure bright copper-foil about 
 a quarter of an inch by half-an-inch in size, and keep 
 the solution gently boiling for forty-five minutes. If 
 at the end of that time the copper remains bright and 
 red, the beer is free from arsenic. 
 
 ,; If a deposit is obtained on the copper, the foil is to 
 be washed successively with water, alcohol, and ether 
 (care being taken that these are pure), dried at a tem- 
 perature not exceeding ioo : C, and subjected to slow 
 sublimation in a thin reduction-tube of small section, 
 and not less than two inches long, the upper portion 
 of which should be warmed before the sublimation 
 begins. For the purpose of the sublimation a small 
 spirit-lamp flame should be used. If any sublimate be 
 obtained, it must be examined under a magnifying 
 power of 200 diameters. Any sublimate which does 
 not show well-defined octahedral or tetrahedral crystals 
 is not to be considered arsenical. 
 
 " N.B. — It must be borne in mind that the blacken- 
 ing of the copper or a deposit thereon from the pre- 
 liminary operation does not demonstrate the presence 
 of arsenic in beer. Abundant blackening and deposit 
 may be obtained from the purest beer." 
 
 Eeading this report a chemist might be tempted to 
 imagine that the real object of the commission of 
 experts was to burke inquiry and to prevent the 
 finding of any arsenic that the beer might have con- 
 tained. Restrict vonrself to Reinsch's test, and 
 
DETECTION AND MEASUREMENT OF ARSENIC. 67 
 
 (notwithstanding that Reinsch, when he was alive, 
 warned chemists that the arsenic existing in the form 
 of arsenic acid required 400 c. c. of strong hydro- 
 chloric acid) take only 30 c. c. of acid with the 200 
 c. c. of beer — that was the recommendation of this 
 commission in the year 1900. 
 
 Among chemists, certainly among chemists as they 
 were thirty or forty years ago, it was customary to 
 control and verify chemical analysis, by resorting to 
 more than one method if that were practicable ; but 
 these experts of the Manchester Brewers' Central 
 Association set their faces against anything of that 
 kind in this instance. They " recommend that the 
 Eeinsch test should be used in preference to all others 
 at present known." In 1844 Fownes placed Marsh's 
 test first, and the copper process of Reinsch in a quite 
 subordinate place. In 1900 a commission of experts 
 restricts itself to the Reinsch test. Verily, in official 
 England that part of chemical analysis which is con- 
 cerned with arsenic has undergone a process of de- 
 generation since Fownes wrote his manual in the 
 year 1844. 
 
 One of the curious features of this report — and 
 that feature has attracted public attention in various 
 quarters — is the complete absence of any quantitative 
 result. The experts reiterate the statement that the 
 sulphuric acid used by Bostock & Company of Liver- 
 pool was arsenical, and that Bostock's sugar became 
 arsenical, and that beer brewed from Bostock's sugar 
 
68 ARSENIC. 
 
 was arsenical, but they do not disclose the degree of 
 the arsenical contamination. AYhy is that ? There is 
 one answer (and it would be a perfectly honest answer, 
 and as appears to the author of this book, it is the only 
 possible honest answer under all the circumstances). 
 That answer is that they themselves had followed the 
 advice given in their report. Their chemical examina- 
 tion was restricted to the use of Eeinsch's test, and 
 Reinsch's test is not quantitative, and, therefore, they 
 had no quantitative results. They did not know the 
 degree of the arsenical contamination either of acid or 
 sugar or beer. 
 
 In the opinion of the author of this book such an 
 investigation as would have given quantitative results 
 was possible if the experts had availed themselves of 
 the resources of analytical chemistry as understood in 
 France and Germany. 
 
 Section 3. — General Testing and Estimation of Arsenic. 
 The almost universal diffusion of arsenic to which 
 attention was called in Chapter I, page 15, constitutes 
 one of the chief difficulties with which the chemist has 
 to contend in all cases of the analysis of materials which 
 contain or are suspected of containing arsenic ; and 
 the necessity of making sure that nothing employed 
 in the analysis contains arsenic must be obvious. All 
 the reagents should be specially tested for arsenic, 
 and for that purpose the Marsh's apparatus is most 
 useful. 
 
DETECTION AND MEASUREMENT OF ARSENIC. 69 
 
 In carrying oat these testings it is to be recom- 
 mended that method be insisted upon ; and that a 
 quantitative character should be given to the results. 
 Thus, for instance, in testing the hydrochloric acid a 
 given volume of acid should be taken, say 50 c. c, 
 and a given quantity of zinc dissolved in the Marsh's 
 apparatus, and the result recorded that there is, or 
 that there is not, the first indication of a trace of 
 arsenic. 
 
 One of the reagents used in estimating arsenic is 
 sulphuretted hydrogen, and in Watts's Dictionary of 
 Chemistry, vol. i. page 367, the following passage 
 occurs : " Sulphuretted hydrogen, evolved by the 
 action of dilute sulphuric acid on sulphide of iron, 
 often contains arsenic proceeding from one or both 
 of the substances used." 
 
 In order to make sure of the purity of the sul- 
 phuretted hydrogen the following procedure might be 
 resorted to : — 
 
 A given weight of the sulphuret of iron being 
 placed in the generating apparatus, a measured volume 
 of diluted hydrochloric acid or diluted sulphuric acid 
 should be gradually poured into the apparatus, and 
 the sulphuretted hydrogen passed into a small Woulff's 
 bottle containing excess of solution of caustic potash 
 or soda, which would absorb the sulphuretted hydro- 
 gen but not the free hydrogen gas, which generally 
 accompanies it, and not the arseniuretted hydrogen 
 if any should be present. The Woulff's bottle should 
 
/O AftSENIC. 
 
 be provided with two necks, through one of which the 
 tube conveying the sulphuretted hydrogen enters, and 
 the other should give exit to another tube to convey 
 any unabsorbed gas and deliver it into another vessel 
 charged with a solution of nitrate of silver. In that 
 way any arsenic existing in the sulphuretted hydrogen 
 would be sure of detection. 
 
 When the passage of the sulphuretted hydrogen is 
 complete, the vessel containing the silver-solution 
 might be detached and dilute ammonia very carefully 
 added to the silver-solution (previously filtered if 
 necessary), in order to show the presence or absence 
 of yellow arsenite of silver. Finally the potash 
 solution which has absorbed the sulphuretted hydrogen 
 might be supersaturated with hydrochloric acid, in order 
 to disclose the presence of the yellow sulphuret of 
 arsenic, if by any chance any arsenic had been absorbed. 
 
 Sulphuretted hydrogen is so important a reagent 
 in the general scheme of the analysis of substances 
 containing metals that chemists require a ready way 
 of preparing it in a condition in which its purity from 
 arsenic is above suspicion. Obviously the solution of 
 sulphuretted hydrogen in caustic potash or soda 
 supplies this want. This solution being placed in an 
 ordinary gas-generating bottle, requires only to be 
 cautiously treated with hydrochloric acid in order to 
 evolve sulphuretted hydrogen gas free from arsenic, 
 inasmuch as any arsenic would remain behind in the 
 state of sulphuret in the gas-generating vessel. 
 
DETECTION AND MEASUREMENT OF ARSENIC. 7 1 
 
 One of the best methods of measuring the amount 
 of arsenic, or estimating as it is termed, is by con- 
 verting the arsenic into the sulphuret of arsenic and 
 weighing the sulphuret. The percentage composition 
 of sulphuret of arsenic As 2 S 3 is as follows : — 
 
 As 2 . . 150 69.72 
 
 S 3 . . 96 30.28 
 
 246 
 
 The factor to multiply by, in order to calculate the 
 amount of arsenic from the weight of the sulphuret, is 
 therefore 0.6972. 
 
 In the recent epidemic of suspected arsenical poison- 
 ing in the North of England, where the root of the 
 mischief is said to have been the arsenical sulphuric 
 acid employed in the transformation of starch into 
 sugar, one of the best methods of analysing the sul- 
 phuric acid would have been by the conversion of the 
 arsenic into sulphuret of arsenic, and the weighing of 
 that sulphuret. According to one of the reports which 
 has found its way into the public press, some of that 
 sulphuric acid contained as much as about 2 per cent, 
 of arsenic. A convenient quantity of such acid would 
 be about 10 grammes. If we suppose that cjuantity 
 to have been weighed out, the steps of the analysis 
 would be as follows : First it would be well to dilute 
 the acid with 50 c. c. of distilled water, and if neces- 
 sary to filter through a small filter, carefully washing 
 the filter. The filtrate would then receive a little 
 
J 2 ARSENIC. 
 
 sulphurous acid, which might be made by heating 
 sulphuric acid and wood charcoal in a very small 
 flask-retort provided with a gas-delivery tube. The 
 gaseous sulphurous acid would be passed into the 
 liquid to be analysed, until there was a strong smell 
 of sulphurous acid ; then the evolution of S0 2 would 
 be stopped and the liquid would be heated to boiling, 
 and so the arsenic, if originally existing as arsenic 
 acid, would be brought into the condition of arsenious 
 acid, and the excess of unacted upon sulphurous acid 
 would be driven off. The next step would be passage 
 of sulphuretted hydrogen into the liquid, and the 
 leaving for a while of the liquid charged w T ith the H 2 S, 
 and corked up so as to exclude the atmosphere as 
 much as possible. In that manner the arsenic origi- 
 nally in the sulphuric acid is made to yield the bright 
 yellow T sulphuret of arsenic, which would next be sepa- 
 rated from the liquid by filtration, and washed with 
 cold water. Probably the sulphuret of arsenic would 
 be mixed with more or less sulphuret of lead, and the 
 colour of the precipitate would be dark and brownish 
 owing to the presence of the lead. A separation of 
 the sulphuret of arsenic from the sulphuret of lead is 
 easily effected by means of a solution of carbonate 
 of ammonia, which dissolves the As 9 S 3 and leaves the 
 sulphuret of lead and other sulphurets as an insoluble 
 black powder. The solution in the carbonate of ammo- 
 nia may be decomposed with dilute hydrochloric acid 
 and the As 2 S 3 reprecipitated, filtered, washed, dried, 
 
DETECTION AND MEASUREMENT OF ARSENIC. 73 
 
 and weighed. Assuming the correctness of the report 
 that there was 2 per cent, of arsenic in the original 
 acid, the weight of the sulphuret of arsenic from the 
 ten grammes of acid would be 0.2869 gramme. 
 
 The sulphuret of arsenic might be mixed with char- 
 coal and carbonate of soda, and the mixture heated to 
 redness in a tube of hard glass, and in that manner 
 made to give a mirror of metallic arsenic ; but some 
 of the arsenic obstinately refuses to become reduced 
 under these circumstances. 
 
 Another way of dealing with the sulphuret is to 
 oxidise it with nitric acid or aqua regia, and convert 
 it into arsenic acid and sulphuric acid, and then to 
 make the ammonio-arseniate of magnesia, which, on 
 gentle ignition in a porcelain crucible, becomes pyro- 
 arseniate of magnesia, which may be weighed. The 
 composition of the pyroarseniate of magnesia is : — 
 
 As., . . -15° 48-39 
 
 Mg 2 . 48 15.48 
 
 7 .... 112 36.13 
 
 10 
 
 and the factor for the calculation of the amount of 
 arsenic from the weight of the pyro-arseniate of mag- 
 nesia is 0.4839. 
 
 The other forms in which arsenic may be advan- 
 tageously weighed are the metallic state and in the 
 form of arsenious acid. 
 
 In the absence of sulphur, an arsenite or an arseniate 
 
74 ARSENIC. 
 
 of potash or soda is totally reduced to the metallic 
 state when it is mixed with a sufficiency of charcoal 
 and heated to redness, and if the operation be per- 
 formed in a narrow tube of hard glass closed at one 
 end, and several inches in length and open at the 
 other end, the metallic arsenic will form a mirror on 
 the inside of the tube. After the tube has cooled, 
 and with the metallic arsenic adherent to the inside, it 
 may be weighed, and then the arsenic may be volati- 
 lised and the empty tube again weighed. The differ- 
 ence between the two weighings is the weight of the 
 metallic arsenic. When sulphur is present, difficulties 
 arise owing to the formation of sulphur salts — arseni- 
 ates wherein the oxygen has been replaced by sulphur 
 — which possess very great stability. Under these 
 conditions there is very great difficulty in effecting 
 a complete volatilisation of every trace of arsenic. 
 The remedy is to take care that there is a considerable 
 excess of alkali or alkaline carbonate, and to pass a 
 slow current of air over the red-hot mass. 
 
 Incidentally the remark, that commercial copper 
 contained more or less arsenic, was cjuoted on page 61. 
 The reader will be prepared to learn that traces of 
 arsenic, and often more than traces of arsenic, are 
 most obstinately retained by some of the heavy 
 metals. 
 
 Trustworthy analyses are on record of some of the 
 best English copper, showing as much as ten grains of 
 arsenic in the pound of copper : and Mexican copper 
 
DETECTION AND MEASUREMENT OF ARSENIC. 75 
 
 has been known to contain two or three per cent, of 
 arsenic. 
 
 When lead is heated with arsenic it forms a brittle 
 alloy, greyish-white in colour, and very fusible. On 
 being strongly heated much of the arsenic is driven off, 
 but a certain proportion of arsenic remains behind with 
 the lead, and is not capable of being driven off. Shot 
 consists of an alloy of lead and arsenic, which contains 
 1.7 per cent, of arsenic. 
 
 The correct measurement of these minute proportions 
 of arsenic in masses of the heavy metals is often a very 
 difficult task, demanding great skill on the part of the 
 analyst. 
 
 Section 4. — Special Organo-Mctallic Condition of 
 Arsenic, &c. 
 
 Passing on to that which more particularly falls 
 within the scope of this book, viz. the poisoning of 
 articles of food or beverages by arsenic, the relations of 
 the arsenic to organic contents will next be considered. 
 
 When arsenic is put into an organic mixture either 
 of two things is possible. The arsenious acid may 
 remain in the mixture in the form of arsenious acid, or 
 its salts, or it may enter into intimate relations with 
 the organic matter as it is peculiarly prone to do. If 
 it remains in the state of arsenious acid, or its salts, 
 it will be recognisable by the ordinary tests, but if 
 it should enter into the peculiarly intimate relation 
 
76 ARSENIC. 
 
 known as the organo-metallic state, then the ordinary 
 tests fail except such as involve destruction of organo- 
 metallic compounds. 
 
 The application of a red heat breaks up organo- 
 metallic combinations, and in all cases where the 
 formation of organo-metallic combinations is suspected 
 the testing for arsenic should be so carried out as to 
 involve such an operation as a preliminary. Neither 
 the usual destructive action of chlorine on the organic 
 matter, nor of a mixture of chlorate of potash and 
 hydrochloric acid, nor of sulphuric acid, quite takes 
 the place of an operation in a combustion-tube, such as 
 is now recommended. 
 
 The chemical history of kakodylic acid, which is a 
 representative of possible organo-metallic arsenical 
 compounds, conveys a warning against trusting any 
 oxidising agent at ordinary temperatures, and nothing 
 short of a furnace operation in the combustion-tube 
 will be satisfactory. 
 
 Failing such an investigation of the sugar and beer 
 which is credited with causing the Manchester epidemic, 
 that epidemic will remain an unsolved mystery. 
 
 The mode of operation to be followed in the instance 
 of suspected arsenic organo-metallic products will now 
 be sketched out. 
 
 The suspected beer may be examined in this 
 manner, the following being the working directions. 
 
 Take ioo c. c. of the beer, add one gramme of pure 
 quicklime slaked with a little water, and evaporate 
 
DETECTION AND MEASUREMENT OF AESENIC. JJ 
 
 down in a clean porcelain dish in the water-bath. 
 When three-quarters of the liquid have evaporated add 
 10 grammes of recently ignited sand, and continue the 
 evaporation down to dryness in the water-bath. The 
 residue (which is a mixture of the dry residue of the 
 beer with the sand and lime) is to be placed in a com- 
 bustion-tube, such as is employed in the well-known 
 elementary organic analysis. 
 
 The combustion-tube is drawn out to a point at one 
 end in the well-known manner, and the point is closed 
 by fusing the glass. The charge placed in the tube is, 
 first, 3 inches of very coarse sand at the closed end ; 
 second, about 9 inches of the mixed lime, sand, and 
 beer residue, and then 3 inches of very coarse sand and 
 an asbestos plug. Then an inch or so in front of the 
 asbestos plug (after the tube has been charged as de- 
 scribed) the tube is to be contracted and drawn out 
 by the aid of the table-blowpipe, and finally so drawn 
 out as to terminate in front in a gas-delivery tube 
 which may be bent and made to deliver gas under 
 water. The arrangement will be perfectly understood 
 by any chemist who has performed a combustion of an 
 organic substance containing mercury. 
 
 The combustion-tube, charged as described, is then 
 placed in the well-known furnace in which organic 
 analyses are carried out, and heated thus : — 
 
 The heat is applied to the tube so that the asbestos 
 ping and three iuches of sand in the front part of the 
 tube are heated to redness, and the closed end of the 
 
78 ARSENIC. 
 
 tube is also heated, and then very gradually the mix- 
 ture of beer residue, &c 3 is made red hot, the heating 
 being from before backwards in the slow and delibe- 
 rate manner which chemists who have made organic 
 analyses will understand. By degrees the beer residue 
 will be converted into a carbonaceous mass mixed with 
 sand and a little lime, and into gases and vapour, for 
 the most part aqueous, which will condense in the 
 front part of the tube, carrying with it any arsenic 
 that may have been contained by the beer. The gas- 
 delivery tube, in which the front of the combustion- 
 tube terminates, should pass into a small quantity of 
 water contained in a very small beaker. 
 
 The combustion-tube, as was explained, is of the 
 usual kind, the hinder part terminating in the usual 
 narrow tail. The proper ending of the operation 
 consists in breaking off the extreme tip of this tail and 
 affixing a narrow india-rubber connecting-tube convey- 
 ing a slow stream of air or oxygen from a small gas- 
 holder, in order to burn up some of the carbonaceous 
 deposit and sweep all traces of arsenic into the small 
 quantity of water, or into the extreme front of the 
 combustion-tube. The small volume of water in the 
 beaker which receives the volatile products of the 
 operation may be treated in many ways. It may be 
 charged with a little chlorine water, so as to convert 
 any traces of arsenic into arsenic acid; and the ulti- 
 mate destiny of the product may be the Marsh's appa- 
 ratus, depending upon the judgment of the chemist. 
 
DETECTION AND MEASUREMENT OF ARSENIC. 79 
 
 Of course a most careful scrutiny of the front part 
 of tbe tube should be made at the end of the operation, 
 and when the tube has cooled down to ordinary 
 temperatures. If any sign of deposit be present it 
 will be necessary to cut the narrow part with a file, 
 and to wash out the front extremity with an appro- 
 priate solvent for traces of arsenic. 
 
 - 
 
 v v or 
 UNIVER3 i 
 
 s£dUF 
 
CHAPTER III. 
 
 THE MANCHESTER EPIDEMIC OF ARSENICAL POISONING 
 AND THE LESSONS TO BE LEARNT FROM IT. 
 
 Manchester, which at the last census had a population 
 of 505,368, and Salford with its population of 198,139, 
 form, in point of fact, one large township; the one 
 occupying the one bank of the narrow river Irwell, 
 and the other occupying the opposite bank of the 
 river. The Manchester epidemic about to be de- 
 scribed was shared by Salford, and the conjoint popu- 
 lation, which at the last census was about 700,000, 
 will no doubt be found to be rather more than that 
 figure when the new census is taken a few weeks 
 hence. 
 
 The following extract from the Manchester Guar- 
 dian of the 23rd November 1900, followed by the 
 reports of the medical officers of Salford and Man- 
 chester, will convey an idea of the epidemic : — 
 
 "An extraordinary state of affairs has just been 
 revealed at some of the hospitals in Manchester and 
 the district. At present the circumstances would 
 appear to poiut to a general epidemic of arsenical 
 poisoning, caused, it is believed, by the consumption 
 
82 ARSENIC. 
 
 of impure beer. The matter is under investigation 
 by medical men who are interested in the cases. The 
 hospital connected with the Manchester Union Work- 
 house at Crumpsall seems to have received a large 
 number of cases. Three mouths ago the Master (Major 
 Ballantine) noticed that the lists of patients from 
 outside the workhouse who had been admitted to the 
 hospital showed a striking inflation when compared 
 with those for the corresponding period of previous 
 years. The patients continued to increase with such 
 rapidity, in fact, that soon there were some 250 more 
 inmates of the hospital than there had been twelve 
 months before. The attention of Dr. Reynolds, the 
 resident medical officer at the workhouse, was drawn 
 to the matter, and that officer commenced a thorough 
 examination of the people under his charge. It 
 seems that Dr. Eeynolds had noticed for some time 
 previously that a large percentage of his patients 
 were apparently suffering from the effects of drink. 
 He now discovered that the common complaint was 
 aggravated in a greater or less degree by arsenical 
 poisoning. Continuing his investigations, he ascer- 
 tained that beer obtained from the same sources as 
 that consumed by the invalids contained traces of 
 arsenic. Dr. Reynolds, along with Dr. Niven (officer 
 of health for Manchester) and other medical men, 
 have, we understand, made a thorough inquiry into 
 the matter, and a report dealing with the whole sub- 
 ject is about to be published. So far as can be 
 
THE MANCHESTER ARSENICAL POISONING. 83 
 
 ascertained, no deaths have occurred which can be 
 directly attributed to the drinking of the beer. The 
 case of the Crumpsall Workhouse, however, is not by 
 any means an isolated one, for the ' complaint/ if it 
 may be so described, is almost equally prevalent in 
 Salford as in Manchester. It is interesting to note 
 that in his weekly return of the health statistics of 
 the city, Dr. Niven, the medical officer of health, 
 states that on the last day of the week ended Novem- 
 ber 17, 1900, there remained in the Manchester 
 Union Infirmary 1099 sick paupers, as compared 
 with 885 at the end of the corresponding week of 
 last year." 
 
 At a special meeting of the Salford Health Com- 
 mittee, held on 26th November 1900, the following 
 report was submitted by the medical officer : — 
 
 " I regret to have to report that there has been a very 
 serious amount of sickness, which is apparently due 
 to arsenical poisoning, in the borough during the past 
 four months. In the third week in July there were 
 three deaths registered in the borough as due to 
 1 peripheral neuritis,' ' multiple neuritis,' or ; alcoholic 
 neuritis.' This was followed week by week by the 
 following numbers up to the present date, viz. : O, I, 
 I, 1, 2, 2, 1, 2, 4, 2, 1, 2, 3, 3, 4, 5, 2, and 2, or a 
 total of 41. In addition to these deaths there have 
 been 25 certified to be due to alcoholism in the same 
 period. To realise exactly the bearing of these figures 
 it should be stated that in the first seven months of 
 
84 ARSENIC. 
 
 the year there were 22 deaths, as compared with 66 
 in the past four months. In the year 1899 there 
 were 19 deaths from multiple neuritis, and 20 from 
 alcoholism. In 1898 the figures were 14 and 17 re- 
 spectively, and in 1897 10 and 17. My attention was 
 naturally called to these figures, but as 20 out of the 
 total cases gave other causes of death, such as pneu- 
 monia, phthisis, thrombosis, influenza, bronchitis, &c., 
 it appeared probable that the cause for the increase 
 was more in the method of certifying than any marked 
 change in the causes of death. Casual inquiries among 
 a few friends confirmed this view, as they had no 
 special amount of sickness of this character in their 
 practices. It must not be forgotten that the whole 
 of these deaths cannot be attributed to the poisoning, 
 for on the average of the deaths from similar causes 
 in the preceding three years there would have been 
 15 deaths in the period under notice, without any 
 disturbing influence, so that this leaves a balance of 
 51 deaths due to the special disease under notice. I 
 have no means of ascertaining the number of cases 
 of sickness, but from the inquiries made it is evidently 
 very great. There have been 100 cases treated by the 
 district Poor-law medical officers, out of which eight 
 have died. 
 
 " On the 9th of November, Dr. Cran, the Poor-law 
 medical officer for the Eegent district, called on me 
 to draw my attention to the large amount of illness 
 of a paralytic nature in his practice, which he had 
 
THE MANCHESTER ARSENICAL POISONING. 85 
 
 concluded was due to beer -drinking. On that day 
 five samples of beer were obtained and sent to the 
 public analyst, and a further ten samples in the few 
 days following. He, however, failed to find anything 
 to account for the illness. Having ascertained that 
 there was a very great number of cases of illuess in 
 the borough of this character, I saw Professor Dele- 
 pine, of the Owens College, and arranged for him to 
 make certain experiments with a view of ascertaining 
 the cause of the prevailing sickness. In the mean- 
 time inquiries were made into the details of all the 
 cases that could be heard of, and it became abundantly 
 evident that the one thing common to them all was 
 beer-drinking, not necessarily to excess, as many of 
 the cases were only those of moderate drinkers ; and 
 it was further ascertained that the beer concerned did 
 not appear to come exclusively from one brewery, but 
 from several. Samples of beer were obtained from a 
 shop to which one case of illness, and that a serious 
 one, could be clearly traced. These were sent to Pro- 
 fessor Delepine, who succeeded in finding arsenic in 
 considerable quantity. In the meantime I ascertained 
 that Dr. Reynolds, to whom the whole credit of first 
 suspecting arsenic as the cause of the illness is due, 
 had also discovered that substance in a sample of beer 
 from another brewery, and that he was of opinion 
 that the illness was arsenical poisoning. The general 
 symptoms of the cases confirm this view ; the paralysis 
 and pigmentation of the skin, running of the eyes and 
 
86 ARSENIC. 
 
 nose, various skin affections, the tenderness of the 
 muscles, and the tingling (' pins-and-needles') sensa- 
 tion in the hands and feet, are all known to be present 
 in arsenical poisoning; but the two symptoms which 
 are generally considered specially characteristic of 
 arsenical poisoning, namely, vomiting and diarrhoea, 
 have been entirely absent in all the cases I have per- 
 sonally examined but one case, where there was some 
 diarrhoea. I was already in communication with the 
 brewers, and I must here thank them most sincerely 
 for the great assistance they have given me in prose- 
 cuting my inquiries, which, but for their help, could 
 not possibly have been carried out so quickly. They 
 threw open their breweries to me without the slightest 
 reserve, and were most anxious that the whole matter 
 should be cleared up and the cause discovered as 
 quickly as possible, although they could not believe 
 that beer could by any possibility be the cause. The 
 sugar and sulphuric acid manufacturers have also been 
 most willing to give me every assistance. I ascer- 
 tained from the brewers that there had not been the 
 slightest change in the method of brewing, or in the 
 materials used, which were in every case purchased 
 from the same firms with whom they had done busi- 
 ness for years. 
 
 "Samples of everything used in the brewery were 
 submitted to analysts, and it was found that arsenic 
 was present in quantity in one sample of glucose and 
 in one sample of inverted sugar. These two sub- 
 
THE MANCHESTER ARSENICAL POISONING. 87 
 
 stances were obtained from the same firm of manu- 
 facturers, and I visited their works, obtaining samples 
 of each ingredient used in the process of manufacture. 
 It should be explained that glucose and inverted sugar, 
 or brewers' sugar, are forms of grape sugar very largely 
 used in brewing, though not, I am led to believe, for 
 cheapening the production. The former is prepared 
 by treating farinaceous substances — in this instance 
 sago, tapioca, and maize meal — with sulphuric acid, 
 and the latter by treating a solution of cane sugar 
 with sulphuric acid. As the arsenic was found in 
 both substances, and the only common ingredient was 
 sulphuric acid, which is well known frequently to 
 contain arsenic, suspicion fastened on this substance. 
 This was confirmed by finding arsenic in large quan- 
 tities in the sulphuric acid from the sugar manufac- 
 turer's works. Before the question can be considered 
 finally settled, however, a great deal of work remains 
 to be done. In the first place, it should be stated that 
 only 2 to 6 per cent, of the affected acid is used in the 
 manufacture of the glucose and sugar, and that only 
 from 5 to 10 per cent, of these materials is used in the 
 beer ; so that careful quantitative analysis, for which 
 there has not yet been time, is required to ascertain if 
 the quantity of arsenic reaching the beer is sufficient 
 to cause the symptoms. It should also be remembered 
 that neither the sulphuric acid manufacturer, the sugar 
 manufacturer, nor the brewer has altered the sources 
 of supply of their raw materials for years, and that, so 
 
88 ARSENIC. 
 
 far as I have been able to ascertain up to the present, 
 there has been no change in any process to account 
 for the change which must have occurred in the beer 
 about June last. 
 
 "At the same time, to have found a poisonous sub- 
 stance in the beer which will account for the symptoms 
 and to have traced that material to its source appears 
 in all probability to have cleared the case up, especially 
 as it has already been found that this firm of manufac- 
 turers supply direct six out of the eight breweries to 
 which cases have been traced, and the same substance 
 is used in breweries in Hey wood and Chester, where 
 cases have also been said to occur. An idea of the 
 poisonous nature of arsenic is best given by stating 
 that one grain in three gallons of beer w ould be a very 
 dangerous amount. It is by no means certain yet that 
 the firm concerned in this case is the only one whose 
 glucose and sugar are contaminated, but as they sell 
 their products to many brewers in this part of Lanca- 
 shire the widespread incidence of the poisoniug is 
 readily explained. There appears to be no donbt that 
 there has been no negligence, much less anything more 
 culpable, on the part of the brewers, and it is satis- 
 factory to have to report that every one concerned 
 has already taken thorough precautions to prevent any 
 further poisoning from this source. I hope to be 
 able to report further on this matter in a short time, 
 giving the results of the quantitative analysis and a 
 detailed account of the cases which have come under 
 
THE MANCHESTER ARSENICAL POISONING. 89 
 
 my notice, also the results of similar analyses of the 
 materials used in the other breweries of the borough. 
 In conclusion, I must thank Professor Delepine for his 
 invaluable assistance and many suggestions ; the fact 
 that such a large number of analyses have been carried 
 out within a week from his receiving the first sample 
 will indicate the enthusiastic manner in which he and 
 his staff have thrown themselves into the work. My 
 thanks are also due to Dr. Cran, Dr. Eeynolds, and 
 other friends for their kind assistance. 
 
 (Signed) " Ch. H. Tattersall, 
 
 Medical Officer of Health, 
 ' ' November 25,1 900. ' ' 
 
 The Manchester Guardian of the 27th November 
 1900 published the following under the heading "Dr. 
 Niven on the Outbreak " : — 
 
 "Dr. Niven, medical officer of health for the city, 
 stated last night that he concurred in the report of Dr. 
 Tattersall, which he regarded as giving a full account 
 of the disease and a satisfactory explanation of its 
 origin. Though the samples of beer which he himself 
 had taken from the sources of supply indicated by 
 sufferers from peripheral neuritis had not been found 
 by the City Analyst to contain arsenic, he was quite 
 satisfied that a considerable quantity of beer on tap in 
 the city was so poisoned, for he had seen the crystal in 
 samples analysed by Professor Delepine. Instances of 
 the disease," continued Dr. Niven, " were being reported 
 
go ARSENIC. 
 
 to him by private practitioners, who had only been 
 able to diagnose it since the publication of Dr. Key- 
 nolds' article. Dr. Niven rather regretted the tendency 
 to ' rush ' and ' exploit ' the epidemic as being inimical 
 to deliberate scientific investigation." 
 
 And, on 3rd December 1900, the epidemic was sub- 
 siding, and the Manchester Guardian announced as 
 follows : — 
 
 "The number of patients suffering from arsenical 
 poisoning in the Manchester Royal Infirmary has now, 
 it is stated, been reduced to ten. Two patients had 
 so far recovered on Saturday that it was possible to 
 remove them to the Barnes Convalescent Home, but 
 during the day a man offered himself for treatment, 
 whose condition was such that it was deemed necessary 
 to detain him in the Infirmary. The number of out- 
 patients suffering from the disease appears to be 
 steadily decreasing. Two patients were, however, 
 admitted on Saturday to the Workhouse Hospital at 
 Withington, where there are a number of cases under 
 treatment. Three new patients were also admitted 
 to the Hope Hospital, but here, as at Withington, 
 none of the cases are thought to be of a serious 
 nature." 
 
 The author of this book has preferred to let the 
 medical officers and, the public press tell the tale of the 
 epidemic ; and the reader will be able to draw his own 
 conclusions. He will no doubt have noted that Dr. 
 Sheridan Delepine, the Professor of Pathology at 
 
THE MANCHESTER ARSENICAL POISONING. 9 1 
 
 Owens College, appeared to give much satisfaction to 
 the medical authorities after the unfortunate Public 
 Analyst had failed to find anything to complain of in 
 fifteen samples of beer, collected some of them on the 
 9th November, and the others "in the few days fol- 
 lowing." After that disappointment, there being a 
 very great number of cases of illness in the borough, 
 the medical officer went to consult Professor Delepine, 
 and "samples of beer were obtained from a shop to 
 which one case of illness, and that a serious one, could 
 be clearly traced. These were sent to Professor Dele- 
 pine, who succeeded in finding arsenic in considerable 
 quantity." "In the meantime," Dr. Tattersall con- 
 tinues, " I ascertained that Dr. Reynolds, to whom the 
 whole credit of first suspecting arsenic as the cause of 
 the illness is due, had also discovered that substance 
 in a sample of beer from another brewery." Very 
 amusing is the latter portion of Dr. Tattersall's report, 
 where he thanks Professor Delepine, and remarks, 
 "The fact that such a large number of analyses have 
 been carried out within a week of his receiving the 
 first sample, will indicate the enthusiastic manner in 
 which he and his staff have thrown themselves into 
 the work." 
 
 In fact, for the purpose of dealing with the epidemic, 
 the Public Analysts of Manchester and Salford seem 
 to have been pushed aside and Professor Delepine and 
 his staff employed in their places. 
 
 Under these circumstances some interest is attachec] 
 
92 AESENIC. 
 
 to a paper published by Professor Delepine, on 12th 
 January 1 901, in the British Medical Journal, in which 
 there is a description of Reinsch's test used "to esti- 
 mate approximately the quantity of the poison." From 
 this pajDer may be quoted : — 
 
 "Eeinsch's Process. — The selection of a proper 
 method for the detection of arsenic in beer was not 
 less important. The two most reliable qualitative 
 methods are undoubtedly the Marsh's and the Reinsch's, 
 but neither of these had been extensively used for the 
 purpose of estimating the amount of arsenic in beer 
 and some of the organic substances we had to deal 
 with. At any rate, references to such experiments 
 were not easy to find. 
 
 " Several comparative trials showed me that the 
 Reinsch's test was eminently applicable to the detec- 
 tion of arsenic in beer, glucose, hops, and all the other 
 organic and inorganic compounds I had to deal with. 
 With regard to Marsh's test it teas inapplicable to beer 
 which had not- been 'previously treated so as to break up 
 the organic matter. I therefore thought that if the 
 first method could be made to reveal sufficiently small 
 quantities of arsenic in beer, and be used to estimate 
 approximately the quantity of the poison in a moderate 
 amount of fluid, it would be preferable to the latter, 
 which, though very delicate, was more complicated, 
 and would probably introduce greater sources of ex- 
 perimental error. By conducting the Reinsch's test 
 with care, I found that it was easy to detect the 
 
THE MANCHESTER ARSENICAL POISONING. 93 
 
 presence of arsenic in ioo c. cm., and even as little 
 as 20 c. cm., of some of the suspected beers. On the 
 other hand, no distinct trace of arsenic con Id be found 
 in as much as 200 c. cm. of beer brewed in Bavaria, 
 where the use of malt substitutes is prohibited by 
 law." 
 
 The passage italicised was not italicised in Dr. Dele- 
 pine's text. It is a curiosity. Marsh, when he was 
 alive, claimed for his process that it was applicable to 
 liquids wherein the organic matter was intact ; and, as 
 the reader may observe on turning back to Marsh's 
 paper, which is republished in extenso, Marsh asserted 
 that he had successfully operated upon " three pints of 
 very thick soup." 
 
 The statement of Dr. Delepine's, that Marsh's test 
 is inapplicable to beer which had not been previously 
 treated so as to break up the organic matter, shows 
 that his knowledge of the test was not derived from 
 Marsh's original paper, but must have come second-hand 
 from some inaccurate text-book. 
 
 Having in that manner disposed of Marsh's test, 
 Professor Delepine describes his own method of work- 
 ing the Keinsch test, quantitatively, as he apparently 
 imagines. He makes a standard solution of arsenious 
 acid in pure beer, containing one part of As 2 3 in one 
 million parts of beer. Then he places his strip of 
 copper in 100 c. c. of the standard solution, and in that 
 way extracts the arsenic from the solution and attaches it 
 to the copper. Next he heats the copper in a peculiar 
 
94 ARSENIC, 
 
 copper vessel and causes the As 2 3 to sublime and 
 form a film on a glass plate which is finally viewed 
 under the microscope. 
 
 The film on this glass plate is to be compared with 
 the films caused by the different samples of beer which 
 he desires to investigate. His own words will best 
 ex{)lain the mode of operation : " To estimate rapidly 
 quantities of As 2 3 above one part per one million it 
 is only necessary to reduce the quantity of beer used 
 until one finds the smallest amount that will give a 
 deposit on copper yielding a sublimate of As 2 3 cor- 
 responding to the standard deposit obtained from 
 ioo c. c. of the xo-owoo- solution. If 25 c. c. of beer 
 are sufficient to give such a result, it may be assumed 
 that there are at least four parts of arsenious acid in 
 one million parts of beer." 
 
 It is further explained that the 25 c. c. of suspected 
 beer might, of course, have been diluted with water 
 and so expanded to give a volume of 100 c. c. be- 
 fore putting the slip of copper-foil into it ; but he 
 thought, and found, that it was better not to dilute 
 in that way. 
 
 Dr. Delepine would seem to be so well satisfied with 
 the results of his work that he did not confine his 
 operations to suspected beer and suspected glucose, 
 but actually took in hand the suspected sulphuric acid 
 reported to be the original fountain of the poisonous 
 outbreak in Manchester and Liverpool. Further on 
 may be read : "By the modification of Eeinsch's method 
 
THE MANCHESTER ARSENICAL POISONING. 95 
 
 just described we found that the amount of arsenious 
 acid in one sample of H 2 S0 4 was about 2 per cent." 
 And there is a pretty little picture of the photograph 
 labelled " Fig. 4. Margin of one of the six films ob- 
 tained from oV of the copper used in testing 2 c. cm. of 
 the sulphuric acid used in the preparation of the above 
 glucose." 
 
 The objections are very serious and far-reaching. 
 
 If the slip of copper were capable of extracting all 
 the arsenic from the beer, and then, on being heated 
 in the little copper receptacle, passed on all the arsenic 
 to the microscope-slides, and if all the arsenic on the 
 slide assumed the crystalline state yielding crystals of 
 equal size, then the method would have claims to be 
 regarded as fairly quantitative. Failure in complying 
 with any of these conditions is fatal to the quantitative 
 character of the method ; and there is failure in two of 
 the conditions. The copper slip does not exhaust the 
 beer and remove all the arsenic. Neither does it 
 always remove the same fraction of the total arsenic 
 in the beer, which might be a practicable substitute 
 for total extraction. 
 
 There is also the greatest uncertainty as to how 
 much or how little of the arsenic reaching the micro- 
 scope-slides will take on the crystalline form. Very 
 slight changes in the physical conditions have much 
 effect upon the production of crystals ; whether the 
 crystals shall be large or small depends upon many 
 other circumstances than the mere quantity of arsenious 
 
g6 ARSENIC. 
 
 acid which happens to be present in the liquid under 
 investigation. 
 
 When the microscope is called in to supplement and 
 complete a process of chemical analysis it brings with 
 it its own peculiar difficulties and limitations. The 
 prettiness of the picture of the slide with microscopic 
 objects upon it is calculated to absorb too much atten- 
 tion, and sometimes so disturbs the judgment that the 
 sense of proportion with the appreciation of numbers 
 suffers from a sort of temporary paralysis. 
 
 The author does not suggest that Dr. Delepine 
 suffers from this disturbance, and indeed assumes that 
 he does not suffer in that way. 
 
 In the paper there is a little picture of the photo- 
 graph from the slide showiug " Sublimate obtained 
 from half of the copper used in testing 50 c. cm. 
 beer containing one part of As 2 3 in one million parts 
 of beer." 
 
 The paper does not give all the numerical data 
 required for a calculation of the weight of the yield 
 of arsenical crystals in the film arising from the heated 
 copper, but the requisite data are most probably in the 
 possession of Dr. Delepine, who no doubt could make 
 the calculation if he were disposed to take the trouble 
 to do so. 
 
 The result of such calculations would show that 
 only a trifling percentage of the As 2 3 put into the 
 standard arsenical beer is ultimately to be found in 
 the film on the glass plate in the form of crystals 
 
THE MANCHESTER ARSENICAL POISONING. 97 
 
 of As 2 3 , the greater portion of the arsenic being 
 either not deposited at all, or else deposited in the 
 amorphous state. 
 
 Keference to Chapter I., pages 16 and 17 (where 
 the peculiarities of the allotropic states of As 2 3 are 
 mentioned), will suggest to the reader how very seri- 
 ously any attempt to gauge the quantity of As 2 3 by 
 the apparent magnitude of the crystals must be inter- 
 fered with. What possible value can be attached to 
 the quantity of the crystals of As 2 3 if the operator 
 is in doubt whether 5 per cent, or 95 per cent, of the 
 As 2 3 on the plate will assume the crystalline form ? 
 The numerical results of these attempts at quantitative 
 valuation by these so-called improvements of Reinsch's 
 test are mere nonsense figures. 
 
 In Chapter II., page 58 and following pages, this 
 question is further dealt with. And the reader will 
 understand that, in the opinion of the author of this 
 book, the chemical investigations into the circum- 
 stances attending this epidemic have been of the 
 most unsatisfactory character, and utterly unworthy 
 of the country of which James Marsh was a native. 
 The actual facts brought out by the investigation 
 into the nature and cause of the epidemic are, as 
 yet, very limited. 
 
 One fact has come out with great distinctness, viz. 
 that not always has the sulphuric acid employed in 
 the preliminary transformation of starch and cane 
 sugar been the acid which is approximately free from 
 
98 ARSENIC. 
 
 arsenic. Whether, however, the acid has ever con- 
 tained anything like 2 per cent, of arsenious acid, or 
 of arsenic in other forms, is, no doubt, within the 
 knowledge of some persons in the North of England ; 
 but no satisfactory evidence has, so far as the present 
 writer knows, come before the public. 
 
 Incidentally, as a kind of by-result of the epidemic, 
 the important observation of Dr. Eeynolds deserves 
 notice. That much of that which has heretofore been 
 set down as chronic alcoholism is in reality chronic 
 poisoning with arsenic has been proved almost to 
 demonstration by Dr. Eeynolds. 
 
 ^ Bearing in mind the general diffusion of arsenic, 
 it must be obvious that we all take minute doses of 
 arsenic with our daily food ; and if the power of dis- 
 posing of arsenic were taken away from us, we 
 should in course of time begin to suffer from arsenical 
 poisoning. In that manner it may be explained how 
 indirectly an excessive drinker might suffer from 
 arsenical poisoning, notwithstanding that his drink 
 might be "practically non- arsenical." The whole 
 question of poisoning with arsenic is a wide and 
 difficult subject. Can the arsenic be stored up in 
 our bones? Can it to any extent replace the phos- 
 phorus in the phosphate of lime? Do the Styrian 
 arsenic-eaters acquire arsenical bones, or do they con- 
 tinually throw off the arsenic in some non-poisonous 
 shape ? — these are questions which have to be answered. 
 The assimilation of mineral matter has at different 
 
THE MANCHESTER ARSENICAL POISONING. 99 
 
 times attracted the attention of chemists, and in this 
 connection the author calls to mind a paper read 
 to the Physiologial Section of the British Associa- 
 tion in the year 1875 by his colleague, W. J. Cooper, 
 wherein questions of this kind were discussed. 
 
 THE END. 
 
 Printed by Ballantyne, Hanson &> Co. 
 Edinburgh &*_ London 
 
RETURN PUBLIC HEALTH LIBRARY 
 
 TO— ► 42 Warren Hall 
 
 642-2511 
 
 LOAN PERIOD 1 
 7 DAYS 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS 
 ALL JOURNALS ARE NON-RENEWABLE 
 
 Return to desk from which borrowed 
 
 DUE AS STAMPED BELOW 
 
 MAY 12 1984 
 
 
 
 REC. D u'3L may i? 
 
 '84 
 
 
 JAN 311990 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 UNIVfc. 
 
 FORM NO. DD26-7, 9m. 1 1 /78 
 
f 
 
 gUSUC HEALTH LIBRAE? 
 
 U.C. BERKELEY LIBRARIES 
 
 CQ2TmbbE5