MEMCAL SCHOOL 
 
 And Pr 
 
Pharmacist, 
 
 .No Dr 
 
 \m>^>ftt; 
 
 \ '■■■ 
 
 '^. 
 
 )i'\±. 
 
 y^^ N. W. Cor. 
 
 Larkin and 
 
 Ellis Sts., San Francisco. ^A 
 
Digitized by the Internet Arciiive 
 
 in 2007 with funding from 
 
 IVIicrosoft Corporation 
 
 http://www.archive.prg/details/compendmanuchemOOeliorich 
 
THE COMPENDIOUS MANUAL 
 
 Qualitative Chemical Analysis 
 
 OP 
 
 C. W. ELIOT AND F. H. STORER 
 
 AS REVISED BY W. R. NICHOLS 
 
 SIXTEENTH EDITION 
 
 NEWLY REVISED 
 By W. B. LINDSAY, A.B., B.S. 
 
 Professor or Chbmistby in Dickinson Colliob 
 
 NEW YORK 
 D. VAN NOSTRAND COMPANY 
 
 23 Murray Street and 27 Warren Street 
 
 1892 
 
 1^ 
 
Copyright, 1891, 
 Bt O. W. ELIOT, F. H. STOKER, and F. W. NICHOLS. 
 
 Ttpoorapht bt J. S. CusHiNQ & Co., Boston. 
 
D 
 
 PREFACE. 
 
 The authors have endeavored to inchide in this short treatise 
 enough of the theory and practice of qualitative analysis " in the 
 wet way," to bring out all the reasoning involved in the subject, 
 and to give the student a firm hold upon the general principles and 
 methods of the art. It has been their aim to give only so much 
 of mechanical detail as is essential to an exact comprehension of 
 the methods and to success in the actual experiments. Hence, the 
 multiplication of different tests or processes, having essentially the 
 same object, has been purposely avoided. For the same reason none 
 of the rare elements are alluded to. The manual is intended to 
 meet the wants of the general student, to whom the study is chiefly 
 valuable as a means of mental discipline and as a compact example 
 of the scientific method of arriving at truth. To professional stu- 
 dents who wish to make themselves expert analysts, this little book 
 offers a logical introduction to the subject, an outline which is 
 trustworthy as far as it goes, but which needs to be filled in and 
 enlarged by the subsequent use of some more elaborate treatise as a 
 book of reference. Professor Johnson, of Yale, has supplied this need 
 with his excellent edition of Fresenius's comprehensive manual. 
 
 The authors believe that they have put into the following pages 
 as much of inorganic qualitative analysis as is useful for training, 
 and also as much as the engineer, physician, agriculturist or liber- 
 ally educated man needs to know. The book has been written for 
 the use of classes in the Institute of Technology, who have already 
 studied the authors' "Manual of Inorganic Chemistry." It is 
 simply an implement devised to facilitate the giving of thorough 
 instruction to large classes in the laboratory. It is the authors' 
 practice to examine their classes orally every four or five exercises, 
 in order to secure close attention to the reasoning of the subject. 
 
 iii 
 
iv PREFACE. 
 
 With this exception, the subject is studied exclusively in the labo- 
 ratory, tools ill hand. Fifty laboratory exercises of two hours each 
 have proved sufficient to give their classes a mastery of the subject 
 as it is presented in this manual. 
 
 It is scarcely necessary to say that this little work is a compila- 
 tion from well-known authorities, among which may be particularly 
 mentioned the works of Galloway, Will, Fresenius, and Northcote 
 & Church. 
 
 Boston, April, 1869. 
 
 PREFACE TO THE REVISED EDITION. 
 
 In this revised edition, undertaken with the advice and consent 
 of the authors, such alterations and additions have been made as 
 have been suggested by the use of the book with a number of 
 
 classes in the laboratory. 
 
 W. R. NICHOLS. 
 BosToy, July, 1876. 
 
 PREFACE TO THE PRESENT EDITION. 
 
 This edition has been carefully revised with the co-operation of 
 
 Professor F. H. Storer. The alterations and additions are such as 
 
 an experience of several years' use of the book has suggested, and 
 
 it is hoped will add to its utility. 
 
 W. B. LINDSAY. 
 
 Dickinson Collbge, 
 Carlisle, Penn., November, 1891. 
 
TABLE OF CONTENTS. 
 
 PAGB 
 
 Introduction. Qualitative analysis defined. Scope of this 
 
 manual. Identifying compounds. Division of the subject. . 1-4 
 
 PART I. 
 
 Chapter I. Example of the separation of two elements. Di- 
 vision of twenty-four metallic elements into seven classes. . . 5-18 
 
 Chapter II, Class I. Chlorides insoluble in water and acids. 
 
 Lead. Silver. Mercury 19-22 
 
 Chapter III. Class II. Sulphides insoluble in water, dilute 
 acids and alkalies. Mercury. Lead. Bismuth. Copper. 
 Cadmium. The precipitation of Classes II and III 23-31 
 
 Chapter IV. Class III. Sulphides insoluble in water or dilute 
 acids, but soluble in alkaline solutions. Arsenic. Anti- 
 mony. Tin. Gold and platinum. Separation of Classes 
 II and III 32-45 
 
 Chapter V. Class IV. Hydrates insoluble in water, ammo- 
 nia-water and solutions of ammonium salts. Simultaneous 
 precipitation of some salts which require an acid solvent. 
 Treatment of the precipitate produced by ammonia-water. 
 Chromium. Aluminum. Manganese. Iron. Separation 
 of Class IV from Classes II and III. The original condition 
 of iron. The use of chloride of ammonium. Interference of 
 organic matter 46-57 
 
 Chapter VI. Class V. Sulphides insoluble in water and in 
 saline or alkaline solutions. Manganese. Zinc. Nickel. 
 Cobalt. Separation of Class V from Class IV 68-64 
 
 V 
 
VI CONTENTS. 
 
 FAOE 
 
 Chapter VII. Class VI. Carbonates insoluble in water, 
 ammonia-water and saline solutions. Barium. Strontium. 
 Calcium. Separation of Class VI from the preceding 
 classes 65-70 
 
 Chapter VIII. Class VII. Three common metallic elements 
 not comprised in the preceding classes. Magnesium. 
 Sodium. Potassium. Table for the separation of the 
 seven classes of the metallic elements 71-75 
 
 Chapter IX. General tests for the non-metallic elements. 
 The classes of salts treated of. General reactions for acids. 
 Metallic elements to be first detected. The barium test. 
 The calcium test. The silver test. Nitrates, chlorates 
 and acetates 76-86 
 
 Chapter X. Special tests for the non-metallic elements. 
 Effervescence. Carbonates, Cyanides. Sulphides. Sul- 
 phites. Hyposulphites (Thiosulphates). Chromates. Ar- 
 senites and Arseniates. Sulphates. Phosphates. Oxalates. 
 Borates. Silicates. Fluorides. Chlorides. Bromides. 
 Iodides. Nitrates. Chlorates. Acetates. Oxides and 
 Hydrates 87-104 
 
 PART II. 
 
 Treatment of Substances of Unknown Composition. Gen- 
 eral observations. Husbanding material 105 
 
 Chapter XI. Treatment of a salt, mineral or other non- 
 metallic solid. Order of procedure. . . . Preliminary ex- 
 amination in the dry way. Closed-tube test. Gases or 
 vapors to be recognized. Sublimates. Reduction test. 
 Metallic globules. . . . Dissolving a salt, mineral or other 
 non-metallic solid, free from organic matter. Dissolving 
 in water. Dissolving in acids. . . . Treatment of solutions 
 obtained. An aqueous solution : neutral, acid, alkaline. 
 An acid solution. . . . Examination of the solutions for 
 the non-metallic elements. . . . Table of solubilities. . . . 
 Treatment of insoluble substances. Fusions. Treatment 
 of the fused mass. Decomposition by means of carbonate 
 of calcium and chloride of ammonium. Fusion with acid 
 sulphate of sodium. Deflagration 106-140 
 
CONTENTS. Vll 
 
 Chapter XII. Treatment of a pure metal or alloy. Action 
 
 of nitric acid on metals. Gold test. Platinum test 141-144 
 
 Chapter XIII. Treatment of liquid substances. Evapora- 
 tion test. Testing with litmus. Testing for ammonia. ... 146, 146 
 
 APPENDIX. 
 
 Reagents. Acids. Sulphuretted hydrogen. Ammonia-water. 
 Ammonium salts. Sodium hydrate. Sodium salts. Potas- 
 sium salts. Iodide of potassium and starch papers. Nitrate 
 of silver. Slaked lime. Lime-water. Calcium chloride. 
 Barium salts. Acetate of lead. Lead paper. Magnesium 
 solution. Ferric chloride. Nitrate of cobalt. Sulphate 
 of copper. Stannous chloride. Oxide of manganese. 
 Mercury salts. Platinic chloride. Zinc. Solution of 
 indigo. Litmus paper. Starch. Alcohol. Water. Hypo- 
 chlorite of sodium. Bisulphide of carbon. Chlorine 
 water 149-158 
 
 Solutions of Known Composition 169-161 
 
 Utensils. Reagent-bottles. Test-tubes. Test-tube rack. 
 Flasks. Beakers. Glass funnels. Filtering. Filter-stand. 
 Rapid filtration. Porcelain dishes and crucibles. Lamps. 
 Blast-lamps and blowers. Iron-stand. Tripod. Wire- 
 gauze. Triangle. Water-bath. Sand-bath. Blowpipes. 
 Platinum foil and wire. Pincers. Platinum crucibles. 
 Wash-bottle. Glass tubing. Stirring-rods. Cutting and 
 cracking glass. Bending and closing glass tubes. Blowing 
 bulbs. Caoutchouc. Corks. Gas-bottle. Self-regulating 
 gas-generator. Mortars. Spatulae 162-197 
 
QUALITATIVE ANALYSIS. 
 
 INTRODUCTION. 
 
 1. Qualitative Analysis, in the widest sense of the term, 
 is the art of finding out the elements contained in com- 
 pound substances. This general definition has important 
 limitations in practice. In the first place, the art, as com- 
 monly taught, applies almost exclusively to mineral, or in- 
 organic, substances, and touches only incidentally upon the 
 multifarious compounds of carbon with hydrogen, oxygen, 
 nitrogen and a few other elements, which form the subject- 
 matter of that branch of chemical science called organic 
 chemistry. Again, the analysis of gases constitutes a dis- 
 tinct branch of analysis, requiring methods and apparatus 
 of its own, and therefore to be most advantageously studied 
 by itself. These deductions made, there remains the analy- 
 sis of inorganic solids and liquids, which is in fact the main 
 subject of qualitative analysis in the present technical sense 
 of the term. 
 
 Only the more important chemical elements are embraced 
 in the systematic course of this manual. Means of detect- 
 ing a few less common elements are incidentally given. 
 Those of the elements which are so rare as to be at present 
 of little interest except to the professional chemist or min- 
 eralogist are not alluded to. 
 
 1 
 
2 IDENTIFYING COMPOUNDS. [§§2,3. 
 
 2. Some previous knowledge of general chemistry is es- 
 sential to the successful study of qualitative analysis. It 
 is assumed that the student knows something of the com- 
 mon elements and of their most important, combinations, 
 that he is familiar with the principal laws which govern 
 chemical changes, and that he possesses a certain skill in 
 the simplest manipulations. The tools and operations em- 
 ployed in qualitative analysis are few and simple ; but neat- 
 ness, method in working and a vigilant attention even to 
 the minutest details, are absolutely essential. As the vari- 
 ous substances used or produced in the operations of analy- 
 sis will not be particularly described, the careful student 
 will keep at hand some text-book on general chemistry, to 
 which he can constantly refer to refresh his recollection of 
 the formulae and physical and chemical properties of the 
 substances referred to. It should be observed that it is 
 often very difficult — in fact, impossible in the present state 
 of knowledge — to express in exact equations the involved 
 or obscure reactions which occur in complex mixtures dur- 
 ing the operations of analysis. It is a useful exercise for 
 students to write out in equations the simpler chemical 
 changes which occur in analysis ; but when the attempt is 
 made to put a complex reaction into numerical symbols, 
 the equations are apt to express either more than we know, 
 or less. 
 
 3. Although the detection of the elements contained in 
 compound substances is the ultimate object of analysis, it 
 is only by exception that the elements themselves are iso- 
 lated, and recognized in their uncombined condition. An 
 element is generally recognized through some familiar com- 
 pound, whose apparition proves the presence of all the ele- 
 ments it contains, just as the presence of any word upon 
 this page makes it sure that the letters with which it is 
 spelt are imprinted there. If, as the result of a definite 
 
§ 3, 4.] IDENTIFYING COMPOUNDS. 3 
 
 series of operations upon some unknown body, the hydrated 
 oxide of iron be produced, no iron having been added dur- 
 ing any stage of the process, the proof of the presence of 
 iron in the original body is quite as certain as if the gray 
 metal itself had been extracted from it. If some well- 
 known sulphate, like sulphate of lead, or of barium, for 
 example, result from a series of experiments upon some 
 unknown mineral, it is certain that the mireral contained 
 sulphur ; provided only that no sulphur has been introduced 
 in any of the chemical agents to whose action the mineral 
 has been submitted. 
 
 The compounds through which the elements are recog- 
 nized are necessarily bodies of known appearance, deport- 
 ment and properties. They are, in fact, bodies of various, 
 though always definite, composition ; oxides, sulphides, chlo- 
 rides, sulphates and many other salts, are thus made the 
 means of identifying one or more of the elements which 
 they contain. The object of the analyst is to bring out 
 from the unknown substance, by expeditious processes and 
 under conditions which admit of no doubt as to their testi- 
 mony, these identifying compounds, with whose appearance 
 and qualities he has previously made himself acquainted. 
 As he follows the course of experiments laid down in this 
 manual, the student will gradually acquire, with the aid of 
 frequent references to a text-book of general chemistry, 
 that stock of information concerning the identifying com- 
 pounds which must be always ready for use in his mind, 
 and at the same time he will be made familiar with the 
 character of the methodical processes which secure a 
 prompt and sure testimony to the elementary composition 
 of the substances he examines. 
 
 4. The subject is treated in two parts or divisions, of 
 which the first contains a series of experiments to illustrate 
 a systematic course of examination for substances in solu- 
 
4 IDENTIFYING COMPOUNDS. 
 
 tion, when once tliat solution has been made; while the 
 second treats chiefly of the preliminary examination of 
 solids and the means of bringing them into solution, and 
 indicates the general methods to be pursued in the actual 
 analysis of a substance of unknown composition. 
 
Part First. 
 
 CHAPTER I. 
 
 DIVISION OP THE METALLIC ELEMENTS INTO CLASSES. 
 
 5. Example of the Separation of two Elements. — Put a 
 small crystal of nitrate of silver and a small crystal of sul- 
 phate of copper into a test-tube (Appendix, § 69), and dis- 
 solve them in two teaspoonfuls (a teaspoonful is equal to 
 five cubic centimeters) of water, warming the water at the 
 lamp to facilitate the solution. Add to this solution a 
 few drops of dilute hydrochloric acid (App., § 3). Shake 
 the contents of the tube violently, wait until the curdy pre- 
 cipitate, which the acid produces, has separated from the 
 liquid, and then add one more drop of hydrochloric acid. 
 If this drop produces an additional precipitate, repeat the 
 operation until the new drop of acid produces no change in 
 the partially clarified liquid. Then, and not till then, has 
 all the silver which the original solution contained been 
 precipitated in the form of chloride of silver, an unemployed 
 balance or excess of the reagent, hydrochloric acid, remain- 
 ing in the clear liquid ; this liquid can be readily separated 
 by filtration from the curdy chloride. ^ 
 
 Shake the contents of the test-tube, and transfer them 
 as completely as possible to a filter (App,, § 74), supported 
 in a very small glass funnel (App., § 73), which has been 
 placed in the mouth of the test-tube . With a wash-bottle 
 (App., § 85) rinse into the filter that portion of the precip- 
 itate which has adhered to the sides of the first test-tube. 
 
 5 
 
6 THE TEEM '' CLASS.'' — CLASS L [§§ 5,6. 
 
 When the filtrate has drained completely from the precipi- 
 tate, set the test-tube which has received it aside. Wash 
 tlie precipitate together into the apex of the filter by means 
 of a wash-bottle with a fine outlet; and, in order to wash 
 out the soluble sulphate of copper which adheres to the 
 precipitate, fill the filter full of water two or three times, 
 throwing away this wash-water when it has passed through 
 the filter. 
 
 The complete separation of the silver and copper which 
 were mixed in the original solution is already accomplished; 
 the silver is on the filter in the form of chloride ; the cop- 
 per is in the clear, bluish filtrate. This speedy and effec- 
 tual separation of the two elements is based upon the fact 
 that chloride of silver is insoluble in water and acid liquids, 
 and is, therefore, formed when hydrochloric acid is added 
 to a solution containing a salt of silver; chloride of copper, 
 however, is readily soluble in water and acid liquids, and, 
 even if formed by the addition of hydrochloric acid to a 
 solution of a compound of copper, would fail to manifest 
 itself by appearing as a precipitate. It is, in general, true 
 that whenever, by the addition of a reagent, there can be 
 formed in any solution a compound insoluble in the liquids 
 present, this compound always separates as a precipitate. 
 Such differences of solubility as are illustrated by the case 
 of the chlorides of silver and copper are the chief reliance 
 of the analyst. 
 
 6. Definition of the Term " Class." Class I. — In the forego- 
 ing experiment only two elements have been separated. It 
 ^ight obviously be very difficult, if not impossible, to find 
 a special reagent for every element, which would always 
 precipitate that single element and never any other. Hy- 
 drochloric acid, for example, which precipitates silver so 
 admirably from any solution containing that element, is 
 capable of eliminating two other elements under like condi- 
 tions. The lower chloride of mercury (mercurous chloride 
 
§ 6.] DIVISION INTO CLASSES. v 7 
 
 or calomel) is insoluble in water and weak acids. Chloride 
 of lead is sparingly soluble in cold water, and is still less 
 soluble in water acidulated with hydrochloric acid. The 
 chlorides of the other metallic elements are all soluble in 
 water and acids under the conditions of the analytical pro- 
 cess. 
 
 There are embraced in the scope of this manual twenty- 
 four of the so-called metallic elements, — elements whose 
 hydrates or oxides are said to be basic in their character, 
 and are collectively designated as bases: if hydrochloric 
 acid were added in proper quantity to a solution imagined 
 to contain all these elements, three, and only three, of the 
 twenty -four elements would be precipitated as chlorides. 
 After filtration and washing, a mixture of chloride of 
 silver, chloride of lead and mercurous chloride would re- 
 main upon the filter, and all the other elements would have 
 passed as soluble compounds into the filtrate. Silver, lead 
 and mercury, the three elements thus separated from the 
 rest by this well-marked reaction with hydrochloric acid, 
 constitute a class, the first of several classes into which the 
 metallic elements are divided for the ends of qualitative 
 analysis. Each class is characterized by some clear reac- 
 tion which suffices, when intelligently applied, to separate 
 the members of any one class from the other classes. The 
 chemical agent, by means of which this distinctive reaction 
 is exhibited, is called the general reagent of the class. Thus, 
 hydrochloric acid is the general reagent of the first class. 
 
 This division of the elements into the classes renders it 
 unnecessary to find means of separating each individual 
 element from all the others. In the systematic course of 
 an analysis, the classes are first sought for and separated; 
 afterwards each class is treated by itself for the detection 
 of its individual members. It is an incidental advantage 
 of this division of the elements into classes that, when the 
 absence of any whole class has been proved by the failure 
 
8 -^% DIVISION INTO CLASSES. [§§ 6-8. 
 
 of its peculiar general reagent to produce a precipitate in 
 a solution under examination, it is unnecessary to search 
 further for any member of that class. Much time is thus 
 saved, for it is as easy to prove the absence of a class as 
 of a single element. The full treatment of the first class of 
 elements, comprising, as we have seen, silver, lead and 
 mercury, is the subject of Chapter II. 
 
 7. Experiment to Illustrate the Division of the Metallic 
 Elements into Classes. — We proceed to demonstrate experi- 
 mentally the chemical facts upon which rests the division 
 of the other metallic elements into convenient classes. 
 
 Prepare a complex solution, by mixing together in a 
 small beaker (App., § 72) a small teaspoonful of each of 
 the following solutions (App., § 66), viz.: — chloride of 
 •^copper, "^rsenious oxide in hydrochloric acid, ferrous chlo- 
 ride, chloride of zinc, chloride of calcium, chloride of mag- 
 nesium and chloride of sodium. Dilute the mixture thus 
 prepared with its own bulk of water. Should any turbidity 
 or precipitate appear, add hydrochloric acid, little by little, 
 until the solution becomes clear. This solution is repre- 
 sentative ; it contains at least one member of each of the 
 classes of elements which remain to be defined. It con- 
 tains no member of the first class ; but we may consistently 
 suppose that the members of this class have been previously 
 precipitated, as in the foregoing experiment (§ 5), and that 
 an excess of hydrochloric acid remains in the liquid. 
 
 8. Definition of Classes II and in. — Pass a slow current 
 of sulphuretted hydrogen (App., § 15) from a gas-bottle or 
 self-regulating generator through the acid liquid in the 
 beaker. This operation must be performed either out of 
 doors or in a current of air sufficient to carry the excess of 
 the gas away from the operator. A dense, dark-colored 
 precipitate will immediately appear, and gradually increase 
 in bulk. When the gas has flowed continually for five or 
 ten minutes through the liquid, remove the beaker from 
 
§ 8.] DEFINITION OF CLASSES II AND fIL 9 
 
 the source of the gas (or interrupt the stream of gas if a 
 self -regulating generator be employed), stir the liquid well, 
 and blow out the sulphuretted hydrogen which lies in the 
 beaker. If after the lapse of two or three minutes the 
 liquid smells distinctly of sulphuretted hydrogen, it is 
 saturated with the gas, and it is sure that the reagpnt has 
 done its work. If the liquid does not retain the character-' 
 istic odor, the gas must be again passed through it until 
 the saturation is certainly attained. 
 
 In order to obtain still further assurance of the satura- 
 tion of the liquid, it is often well to take the first portions 
 of the filtrate of the succeeding paragraph and add a small 
 quantity of sulphuretted hydrogen water (App., § 16) or 
 pass sulphuretted hydrogen through it. If a sufficient 
 amount of gas has not been passed into the liquid, the ad- 
 dition of the sulphuretted hydrogen water would cause the 
 appearance of a precipitate. In such a case the filtered 
 portion must be returned to the beaker and the stream of 
 gas again passed through the liquid. 
 
 Pour the contents of the beaker, well stirred up, upon a 
 filter which is supported over a test-tube or second beaker. 
 Rinse the first beaker once with a teaspoonful of water, and 
 transfer this rinsing water to the filter, allowing the filtered 
 liquid to mix with the original filtrate. Label ^ this filtrate 
 
 iThe student should at once make it a rule to label every filtrate or 
 precipitate which he has occasion to set aside, even for a few moments. A 
 bit of paper large enough to carry a descriptive symbol or abbreviation 
 should be attached to the vessel which contains the liquid or precipitate. 
 Paper gummed on the back, or the small labels which are sold already 
 gummed, are convenient for this use. 
 
 This habit, once acquired, will enable the studeut to carry on simul- 
 taneously, without error or confusion, several operations. He may be 
 throwing down one precipitate, washing another, filtering a third and 
 dissolving a fourth at the same time, and the four processes may belong to 
 as many different stages of the analysis. There will be no danger of error 
 if labels are faithfully used ; and a great deal of time will be saved. The 
 unaided memory is incapable of doing such work with that full certainty, 
 admitting of no suspicion or after-qualms of doubt, which is alone satisfy- 
 ing, or indeed admissible, in scientific research. 
 
10 ,. S CLASSES II AND III. [§ 8. 
 
 " Filtrate from II and III " (classes), and preserve it for 
 later study. 
 
 If any considerable quantity of precipitate has adhered 
 to the sides of the original beaker, it may be detached and 
 washed on to the filter by means of a sharp jet of water 
 from the wash-bottle. The precipitate, as it lies upon the 
 filter, must then be washed once or twice with water ; the 
 wash-water is thrown away. The washed precipitate con- 
 sists of a mixture of sulphide of copper (CuS) and trisul- 
 phide of arsenic (As^Sg) . The fact that these sulphides are 
 precipitated under the conditions of this experiment proves 
 that they are both insoluble in weak acid liquors. They 
 are also both insoluble in water. But an important differ- 
 ence between the two sulphides nevertheless exists, a differ- 
 ence which affords a trustworthy means of separating one 
 from the other. 
 
 When the water has drained away from the precipitate, 
 open the filter upon a plate of glass, and gently scrape the 
 precipitate off the paper with a spatula of wood or horn. 
 Place the precipitate in a small porcelain dish (App., § 77), 
 pour over it enough of a solution of sulphide of sodium 
 (App., § 24) to somewhat more than cover it, and heat the 
 mixture cautiously to boiling, stirring it all the time with 
 a glass rod. The quantity of sulphide of sodium to be 
 employed varies, of course, with the bulk of the precipi- 
 tate; in this case two or three teaspoonfuls will probably 
 suffice. It is very undesirable to use an unnecessarily large 
 quantity of the reagent for reasons that will h^eafter ap- 
 pear. A portion of the original precipitate remains undis- 
 solved ; but a portion has passed into solution. Filter the 
 hot liquid again. The black residue on the filter is sul- 
 phide of copper, which is insoluble, not only in water and 
 weak acids, but also in the alkaline sulphide. To the filtrate, 
 collected in a test-tube, add gradually hydrochloric acid, 
 until the liquid has an acid reaction on litmus paper 
 
§8.] MSRCUBT AND LEAD. 11 
 
 (App., § 59). A yellow precipitate of sulphide of arsenic 
 will appear as soon as the alkaline solvent which kept it in 
 solution is destroyed. The sulphide of arsenic differs from 
 the sulphide of copper in that it is soluble in alkaline 
 liquids. 
 
 In this series of experiments copper and arsenic stand, 
 not as isolated elements, but as representatives of classes. 
 The following common elements have sulphides which are 
 insoluble in water, weak acids and alkaline liquids : — 
 Lead, mercury, bismuth, cadmium and copper. These 
 elements constitute Class II in our system of analysis. 
 The following elements have sulphides which are insoluble 
 in water and weak acids, but soluble in alkaline liquids : — 
 Arsenic, antimony, tin and the precious metals gold and 
 platinum. These elements constitute Class III. If all 
 the elements of both groups had been present in the original 
 solution, one class might have been separated from the 
 other by the same process employed in the case of the 
 representative elements, arsenic and copper. 
 
 The question may naturally suggest itself, how it hap- 
 pens that lead and mercury are included in Class II, when 
 they were both precipitated in Class I. The chloride of 
 lead, which is thrown down by hydrochloric acid, is not 
 wholly insoluble in water or dilute hydrochloric acid; hence 
 it happens that the lead is not completely precipitated in 
 Class I. That portion of the lead which has escaped pre- 
 cipitation as chloride in Class I, will be thrown down as 
 sulphide in Class II, for the sulphide of lead is insoluble 
 in water, weak acids and alkalies. In regard to mercur^^ 
 it will.be remembered that there are two sorts of mercury 
 salts, mercurous salts and mercuric salts. The mercurous 
 chloride, HgCl (calomel), is insoluble in water ; but the mer- 
 curic chloride, HgCl.^ (corrosive sublimate), is soluble in water. 
 If, therefore, mercury be present in the form of some mer- 
 curous salt, it will be separated as mercurous chloride in 
 
12 DEFINITION OF CLASS IV. [§§ 8, 9. 
 
 Class I. If, on the contrary, it be present in the form of 
 some mercuric salt, it will be separated in Class II as mer- 
 curic sulphide (HgS), for this sulphide is insoluble in water, 
 weak acids and alkaline liquids. If a mixture of mercurous 
 and mercuric salts be contained in the original solution, 
 mercury will appear both in Class I and in Class II. 
 
 The treatment of Class II is fully discussed in Chapter 
 III. The separation of Class III and the means of separat- 
 ing the members of the class, each from the others, form 
 the subject of Chapter IV. 
 
 9. Definition of Class IV. — We now return to the study 
 of the filtrate from Classes II and III. Pour the liquid 
 into a small evaporating-dish, and boil it gently for five or 
 six minutes to expel the sulphuretted hydrogen with which 
 the fluid is still charged. To make sure that all the gas is 
 expelled, hold a bit of white paper moistened with a solu- 
 tion of acetate of lead (App., § 47) over the boiling liquid; 
 when the paper remains white, all the sulphuretted hydro- 
 gen is gone. Next, add to the liquid in the dish ten or 
 twelve drops of strong nitric acid (App., § 4), and again 
 gently boil the liquid for three or four minutes, in order 
 that all the iron present may be converted into ferric salts. 
 Then pour the liquid into a test-tube, add to it about one 
 third its bulk of chloride of ammonium (App., § 20), and 
 finally add ammonia- water (App., § 17), little by little, until 
 the mixture, after being well shaken, smells decidedly of 
 ammonia. A brownish-red precipitate of ferric hydrate will 
 separate from the liquid. Pour the contents of the test-tube 
 ^on a filter, rinse the tube and the precipitate once with 
 a little water, and preserve the whole filtrate for subsequent 
 operations. 
 
 Two other metals, aluminum and chromium, are precipi- 
 tated, as iron has here been, by ammonia-water under the 
 same conditions and in the same form, viz., as hydrates. 
 These three elements, therefore, constitute the fourth class, 
 
§§ 9, 10.] DEFINITION OF CLASS V. 13 
 
 whose treatment forms the subject of Chapter V. The 
 hydrates of these elements are insoluble in water, even in the 
 presence of salts of ammonium, such as the chloride of am- 
 monium which has been expressly added, and the nitrate of 
 ammonium which has been formed during the neutraliza- 
 tion of the acid liquid. The student may be curious to 
 know why the presence of ammonium salts is insisted upon 
 before the elements of this class are thrown down by am- 
 monia-water, j The ammonium salts have nothing to do 
 with the precipitation of iron, aluminum and chromium; 
 but by their presence they prevent the precipitation, as will 
 be hereafter explained, of certain other elements whose 
 hydrates, though but slightly soluble in water, are dissolved 
 by solutions of ammonium salts. The salts of ammonium 
 are therefore added to keep in solution certain other ele- 
 ments which otherwise would encumber Class IV. \ 
 
 10. Definition of Class V. — We now proceed to the exami- 
 nation of the filtrate from the precipitate of Class IV. 
 Bring this liquid to boiling in a test-tube, and add sulphide 
 of ammonium (App., § 18), little by little, to the boiling 
 liquid as long as a precipitate continues to be formed. To 
 make sure that the precipitation is complete, shake the hot 
 contents of the test-tube strongly, and then allow the mix- 
 ture to settle until the upper portion of the liquid becomes 
 clear. Into this clear portion let fall a drop of sulphide of 
 ammonium ; when this drop produces no additional precipi- 
 tate, the precipitation is complete. Filter off the whitish 
 precipitate of sulphide of zinc, and preserve the filtrate 
 for further treatment. It sometimes happens that this 
 precipitate refuses to settle and leave the upper portion of 
 the liquid sufficiently clear to test in the manner described 
 above ; in this case a small portion of the mixture may be 
 filtered and a drop of sulphide of ammonium added to the 
 clear filtrate in order to determine whether the precipi- 
 tation is complete. If not, the filtered liquid must be 
 returned to the flask and more of the reagent added. 
 
14 LETWITION OF CLASS VI [§§ 10, 11. 
 
 I The element zinc, representing a new class of elements, 
 is precipitated under the conditions of the above experi- 
 ment, because its sulphide, though soluble in dilute acids, 
 is insoluble in alkaline liquids. | The metals manganese, 
 nickel and cobalt resemble zinc in this respect, and these 
 four elements therefore form a new class. Class V, in this 
 analytical method. The representative sulphide of this class 
 was not precipitated by the sulphuretted hydrogen when 
 that reagent was employed to throw down the members of 
 the Classes II and III, because the solution was at that time 
 acid. ; Again it was not precipitated with Class IV by the 
 ammonia-water, because the sulphuretted hydrogen with 
 which the solution had previously been charged, was expelled 
 by boiling before the ammonia-water was added. ) The com- 
 plete treatment of Class V forms the subject of Chapter VI. 
 11. Definition of Class VL — Add to the filtrate from 
 Class V, two or three teaspoonfuls of carbonate of ammo- 
 nium (App., § 19) and boil the solution. A white precipitate 
 of carbonate of calcium will be produced. After boiling, 
 allow the precipitate to settle until the upper portion of the 
 liquid is comparatively clear. To this clarified portion add 
 a fresh drop of carbonate of ammonium. If this drop pro- 
 duce an additional precipitate, more carbonate of ammo- 
 nium must be added, and the boiling repeated. To the 
 partially clarified liquid add again a drop of carbonate of 
 ammonium. This process of making sure of the complete 
 precipitation of the calcium is essentially the same as that 
 prescribed in precipitating the last class, and is, indeed, of 
 general application. When the precipitation of the calcium 
 has been proved to be complete, filter the whole liquid, and 
 receive the filtrate in a small evaporating-dish. Calcium is 
 separated in the form of carbonate under these circum- 
 stances, because this carbonate is almost insoluble in weak 
 alkaline liquids, when an excess of carbonate of ammonium 
 is present. The allied elements barium and strontium 
 
§§ 11, 12.] DEFINITION OF CLASS VII. 15 
 
 behave in the same way, so that these three elements, viz., 
 barium, strontium and calcium, compose a new class — 
 Class VI, whose complete treatment is set forth in Chapter 
 VII. 
 
 12. Definition of Class VII. — Of the twenty-four metallic 
 elements, which were to be classified (§ 6), only three 
 remain, viz., magnesium, sodium and potassium. It is 
 obvious that these three elements could not have remained 
 in solution through all the operations to which the original 
 liquid has been submitted, unless their chlorides and sul- 
 phides had been soluble in weak acids, and their oxides (or 
 hydrates), sulphides and carbonates soluble in dilute ammo- 
 nia-water, at least in presence of dilute solutions of ammo- 
 nium salts. It is a fact that all these compounds of sodium 
 and potassium are soluble in water, and in weak acid, alka- 
 line and saline solutions; the magnesium would have been 
 partially precipitated in Classes IV, V and VI, but for the 
 presence of ammonium salts in the solution. These three 
 elements constitute Class VII. 
 
 Evaporate the filtrate from Class VI until it is reduced 
 to one half or one third of its original bulk. Pour a small 
 part of the evaporated filtrate into a test-tube ; add a little 
 ammonia-water and a teaspoonful of phosphate of sodium 
 (App., § 27), and shake the contents of the tube violently. 
 Sooner or later a crystalline precipitate will appear. This 
 peculiar white precipitate of phosphate of magnesium and 
 ammonium ' identifies magnesium ; but as we have added a 
 reagent containing sodium, the filtrate is useless for further 
 examination. The liquid remaining in the evaporating- 
 dish is then evaporated to dryness, and moderately ignited 
 until fuming ceases. All the ammoniacal sal^ which the 
 solution contained will be driven off by this means, and 
 there will remain a fixed residue, in which are concentrated 
 all the salts of magnesium and sodium which the solution 
 contained. In this case we have already proved the presence 
 
16 SEPARATION OF CLASSES. [§§ 12-14. 
 
 of magnesmm ; it remains to indicate briefly the nature of 
 the means used to detect the sodium. 
 
 Dissolve the residue in the dish, or a portion of it, in 
 three or four drops of water. Dip a clean platinum wire 
 (App., § 83) into this solution, and introduce this wire into 
 the colorless flame of a gas or spirit-lamp (App., § 78). 
 An intense yellow coloration of the flame demonstrates the 
 presence of sodium. A violet coloration would have proved 
 the presence of potassium. Magnesium compounds, when 
 present, have no prejudicial effect on these characteristic 
 colorations. The means of detecting each member of this 
 last class in presence of the others will be found described 
 in Chapter VIII. 
 
 13. A condensed statement of the classification illustrated 
 by the foregoing experiments is contained in the table on 
 the next page. All the common metallic elements are em- 
 braced in it. The classification itself would not be essen- 
 tially different, if all the rare elements were comprehended 
 in it. The general subdivisions would be the same, although 
 some of them would embrace many more particulars. .• 
 
 14. It is essential to success to follow precisely the pre- 
 scribed order in applying the various general reagents. Class 
 I would go down with Class II, were hydrochloric acid for- 
 gotten as the first general reagent. Class II would be pre- 
 cipitated in part with Class IV and in part with Class V if 
 sulphuretted hydrogen were not used in its proper place. 
 A large number of the members of the first 'five classes 
 would be precipitated as carbonates with Class VI, were 
 they not previously eliminated by the systematic application 
 cf hydrochloric acid, sulphuretted hydrogen, ammonia- water 
 and sulphide of ammonium in the precise order and under 
 the exact conditions above described. It should be noticed 
 that all the general reagents are volatile substances, which 
 can be completely removed by an evaporation to dryness 
 followed by a very moderate ignition. 
 
§14.] 
 
 CLASSIFICATION. 
 
 17 
 
 S,fv 
 
 a GirtH</j 
 
 jFi iv*^ 
 
 /i/1/wv*- yyt^* 
 
 
 Precipita- 
 ted as 
 chlorides. 
 
 Ag 
 Pb 
 Hg 
 
 9 
 
 >■ 
 
 09 
 OQ 
 M 
 
 Q2SS 
 
 5 
 
 Precipitated as 
 sulphides insoluble 
 in dilute acids, and 
 not redissolved by 
 
 alkaline fluids. 
 
 Class II. 
 
 5&?g& 
 
 Precipitated as 
 
 sulphides insoluble 
 
 in dilute acids, but 
 
 redissolved by 
 
 alkaline fluids. 
 
 Class III. 
 
 [together with cer- 
 tain salts which re- 
 quire an acid sol- 
 vent] . 
 
 Q 
 
 Precipitated by 
 
 ammonia-water, 
 
 usually as hydrates, 
 
 re 
 
 Al 
 
 t 
 
 < 
 
 ?si! 
 
 Precipitated 
 
 as sulphides 
 
 insoluble in 
 
 alkaline fluids. 
 
 Zn 
 
 o 
 
 ► 
 
 09 
 09 
 
 < 
 
 
 Precipita- 
 ted as car- 
 bonates. 
 
 Ca 
 Ba 
 Br 
 
 a 
 
 >> 
 
 CO 
 OB 
 
 < 
 
 «?i 
 
 Remaining 
 
 elements. 
 
 Distinguished 
 
 by special 
 
 teste. 
 
 o 
 
 < 
 
 H- ( 
 
 O 
 
 It* 
 > 
 
 CD 
 
 CO 
 
 CO 
 
 O 
 M 
 
 f 
 
 CO 
 
18 SEPARATION OF CLASSES. 
 
 15. The series of experiments just completed is merely 
 intended to demonstrate the principles in accordance with 
 which these twenty-four elements are classified for the pur- 
 poses of qualitative analysis. The general plan is here 
 sketched; the practical details, essential to success in the 
 conduct of an actual analysis, will be given hereafter. 
 
tx 
 
 CHAPTER II. ' 
 
 CLASS I. — ELEMENTS WHOSE CHLORIDES ARE INSOLU- 
 BLE IN WATER AND ACIDS. 
 
 16. Example of the Precipitation of the Members of Class 
 I. — Place in a test-tube five or six drops of a tolerably con- 
 centrated aqueous solution of nitrate of silver (App., § 66), 
 an equal quantity of a solution of mercurous nitrate and 
 two teaspoonfuls of a solution of nitrate of lead. In case 
 the solution becomes turbid through the action of carbonic 
 acid dissolved in the water, pour in one or two drops* of 
 nitric acid to destroy the cloudiness. 
 
 Add dilute hydrochloric acid to the solution, drop by 
 drop, and shake the mixture thoroughly after each addi- 
 tion of the acid, until the fresh portions of the latter cease 
 to form any precipitate on coming in contact with the com- 
 paratively clear liquor which floats above the insoluble 
 chlorides. Finally, add three or four more drops of the 
 acid to insure the presence of an excess of it in the solu- 
 tion. 
 
 17. Analysis of the Mixed Chlorides. — The following 
 method of separating the chlorides of lead, silver and mer- 
 cury, one from another, depends upon the facts : — 1st. 
 That chloride of lead, though but little soluble in cold water 
 or dilute hydrochloric acid, dissolves readily in boiling 
 water, while chloride of silver and subchloride of mercury 
 (mercurous chloride, HgCl) are as good as insoluble in that 
 liquid; 2d. That chloride of silver is soluble in ammonia- 
 water; and 3d. That mercurous chloride is discolored by 
 ammonia-water without dissolving. 
 
 19 
 
20 SEPARATION OF LEAD, [§ 17. 
 
 To effect the separation: — Collect upon a filter the pre- 
 cipitate produced by hydrochloric acid, allow it to drain, and 
 rinse it with a few drops of cold water. Place a clean test- 
 tube beneath the funnel which contains the filter and pre- 
 cipitate, thrust a glass rod through the apex of the filter, 
 and wash the precipitate off the filter into the test-tube by 
 means of a wash-bottle which throws a fine stream. 
 
 Heat the mixture of water and precipitate to boiling, 
 then allow the precipitate to settle and pour off the hot 
 liquor upon a new filter, taking care to retain the precipi- 
 tate as far as possible in the tube. To the clear filtrate 
 add two or three teaspoonfuls of dilute sulphuric acid 
 (App., § 10). A white cloud of sulphate of lead will be 
 
 Test formed in the midst of the liquid. In case the 
 
 for precipitate contains a large proportion of chloride 
 
 ^^- of lead, it may happen that the hot water will take 
 up so much of it that crystals of the chloride will separate 
 from the clear aqueous solution as it becomes cold, or that 
 the liquor will be rendered cloudy by the deposition of 
 numerous small particles of the chloride. 
 
 Pour a fresh quantity of water upon the precipitate which 
 was retained in the test-tube, boil the mixture and, after 
 allowing the precipitate to subside, pour the nearly clear 
 liquid upon the same filter as before. This operation of 
 boiling the precipitate with successive portions of water is 
 performed in order to insure the complete removal of the 
 chloride of lead; the liquid is filtered through the same 
 filter in order to retain any particles of the precipitate which 
 fail to settle in the tube, but it is not necessary to save this 
 wash- water as most of the chloride of lead was obtained in 
 the filtrate from the first boiling. 
 
 After the mixed precipitate of chloride of silver and mer- 
 curous chloride has been thus boiled with water several 
 times, cover it with several teaspoonfuls of ammonia-water, 
 heat the mixture to boiling, and pour it upon the filter used 
 
§§ 17, 18.] SILVER AND MEBCUBT, 21 
 
 in the last paragraph : the filtrate is to be received in a clean 
 test-tube. The chloride of silver, dissolved by the ammo- 
 nia-water, will pass into the filtrate, while the mercurous 
 chloride suffers decomposition, and is converted into an ob- 
 scure compound of mercury, chlorine, nitrogen and hydrogen, 
 which remains upon the filter in the form of an insoluble 
 black or gray powder. 
 
 To confirm the presence of silver, add to the filtrate dilute 
 nitric acid (App., § 6), until the solution will red- ^est 
 den litmus paper: the chloride of silver is repre- for 
 cipitated unchanged as soon as the alkaline solvent -^S- 
 is neutralized. 
 
 To confirm the presence of mercury, the metal itself 
 may be set free by heating the dry residue with carbo- 
 nate of sodium (App., § 25) in a glass tube. To insure 
 the success of this experiment, wash into the lowest point 
 of the filter the whole of the black residue. As soon as 
 the last drops of liquid have drained from the filter, dry 
 the latter, either in a dish upon a water-bath, or J>y spread- 
 ing it open upon a ring of the iron stand (App., § 80) placed 
 high above a small flame of the gas-lamp. When the pre- 
 cipitate is completely dry, scrape it from the paper, mix it 
 with an equal bulk of carbonate of sodium pre- ^est 
 viously dried over the gas-lamp on platinum foil for 
 (App., § 83) unless already perfectly dry, and trans- ^6- 
 fer the mixture to the bottom of a glass tube. No. 5 (App., 
 § 86), closed at one end. Then wipe out the inside of the 
 tube with a tuft of cotton fixed to a wire, or with a twisted 
 slip of paper, and heat the closed end of the tube for two 
 or three minutes in the flame of a gas-lamp. A sublimate 
 of finely divided metallic mercury will form upon the walls 
 of the tube ; it will cohere to visible globules when scratched 
 with a piece of iron wire. 
 
 18. An outline of the operations described in the fore- 
 going paragraphs may be presented in tabular form, as fol- 
 lows : — 
 
22 CLASS I. [§§ 18, 19. 
 
 The General Reagent (HCl) of Class I precipitates PbClj, AgCl 
 and HgCl. When the precipitate is boiled with water : — 
 
 PbCla goes into 
 solution. C o n - 
 firm presence of 
 lead by precipita- 
 tion of sulphate 
 of lead. 
 
 AgCl and HgCl remain ' undissolved. On 
 treating the mixture with ammonia- water : — 
 
 AgCl dissolves. 
 
 Confirm presence 
 of silver with nitric 
 acid. 
 
 A black compound of 
 Hg remains undissolved. 
 Confirm presence of Hg 
 by isolating the metal. 
 
 19. In the actual analysis of a solution of unknown com- 
 position, a precipitate might under certain circumstances 
 be formed on the addition of hydrochloric acid, even in the 
 absence of all members of Class I. This might occur in 
 case the liquid under examination contained a hyposulphite 
 (thiosulphate) ; for this class of salts is decomposed, with 
 evolution of sulphurous acid and deposition of sulphur, 
 on the addition of the general reagent (HCl) of Class I. 
 Some sulphides also are decomposed by hydrochloric acid, 
 with deposition of sulphur. An acid solution of antimony, 
 bismuth or tin with some acid other than hydrochloric 
 may give a precipitate of the basic chlorides of these ele- 
 ments: their basic chlorides are soluble in an excess of 
 hydrochloric acid. Concentrated solutions of certain salts 
 such as barium chloride and nitrate, sodium chloride, etc., 
 may give a precipitate of the salt itself, on the addition of 
 hydrochloric acid; such precipitates are readily soluble in a 
 small amount of water. A gelatinous white precipitate of 
 hydrated silicic acid might also be formed at this stage, 
 and other precipitates in certain circumstances, as will be 
 explained het.eafter (§§ 70 and 88, I. C). 
 
CHAPTER III. 
 
 CLASS II. — ELEMENTS WHOSE SULPHIDES ARE INSOLU^ 
 BLE IN WATER, DILUTE ACIDS AND ALKALIES. 
 
 20. Example of the Precipitation of the Members of Class 
 n. — Place in a small beaker a half teaspoonful of a solution 
 of each of the following substances: — mercuric jchlori^e 
 (corrosive sublimate), chloride of bismuth, of cadmiuin and 
 of copper, together with two or three teaspoonfuls of a cold 
 aqueouT solution of c hloride of lead . Pill the beaker half 
 full of water : a white precipitate of the basic chloride of 
 bismuth usually falls on the addition of the water, which 
 may be disregarded. 
 
 Heat the liquid in the beaker nearly or quite to boiling, 
 then place the beaker beneath a chimney or in a strong 
 draught of air, and saturate the solution with sulphuretted 
 hydrogen gas. To determine when enough sulphuretted 
 hydrogen has been passed through the liquid, remove the 
 beaker every four on five minutes from the source of the 
 gas, blow away the gas which lies in the beaker above 
 the liquid, and stir the latter thoroughly with a glass rod. 
 If, after the lapse of two or three minutes, the liquid still 
 smells strongly of sulphuretted hydrogen, it is saturated 
 with the gas and ready to be filtered. But in case no per- 
 sistent odor of sulphuretted hydrogen is observed, the gas 
 must be passed anew through the liquor until it has become 
 fully saturated. Since some of the substances above enu- 
 merated are thrown down more quickly by sulphuretted 
 hydrogen than the others, it is absolutely necessary to em- 
 ploy the reagent in excess in order that those members of 
 
 23 
 
24 SEPARATION OF CLASS II. [§§ 29) 21. 
 
 the class which are least easily precipitated may not escape 
 detection. To make sure that precipitation is complete 
 filter a portion, and to the clear filtrate in a test-tube or 
 small beaker add an equal volume of sulphuretted hydrogen 
 water; if any precipitate appear, the whole of the liquid 
 should be diluted, and sulphuretted hydrogen passed through 
 it again until a portion filtered and tested as before gives 
 no precipitate. If the solution is too strongly acid, there is 
 great danger of incomplete precipitation, especially of cad- 
 mium. 
 
 21. Analysis of the Mixed Sulphides. — The following 
 method of separating the members of Class II depends 
 upon the facts : — 1st. That mercuric sulphide is insoluble 
 in hot dilute nitric acid, while the other sulphides are con- 
 verted thereby into soluble nitrates. 2d. That sulphate of 
 lead is insoluble in acidulated water, while the sulphates 
 of the other -djiembers of the class are soluble. 3d. That 
 hydrate of bismuth is insoluble in ammonia-water, while 
 the hydrates of cadmium and copper are soluble in that 
 liquid. 4th. That sulphide of copper is soluble in solution 
 of cyanide of potassium while sulphide of cadmium is 
 insoluble. 
 
 To effect the separation : — Collect the precipitated sul- 
 phides upon a filter; wash the precipitate thoroughly with 
 successive portions of water until the wash-water is no 
 longer acid to litmus paper; transfer the precipitate to a 
 small parcel^in dish, pour upon it four or five times as 
 much dilute nitric acid as would be sufiicient to cover it, 
 and boil the mixture during two or three minutes, stirring 
 it constantly with a glass rod, and adding water or dilute 
 nitric acid at intervals to replace the liquid which evapo- 
 rates. All the sulphides with the exception of the sulphide 
 of mercury are decomposed and the several elements go into 
 solution as nitrates ; the sulphide of mercury, mixed with 
 some free sulphur resulting from the decomposition of the 
 
§ 21.] SEPARATION OF MERCURY. 25 
 
 other sulphides, remains undissolved as a heavy dark-colored 
 mass. Decant the nitric acid solution into a filter, collect 
 the filtrate in a second porcelain dish, and evaporate it 
 nearly to dryness in order to drive off the greater part of 
 the free nitric acid before examining it for the elements 
 supposed to be contained in it. 
 
 The residue containing mercury, insoluble in dilute nitric 
 acid, may sometimes be light colored if the boiling with 
 nitric acid is too prolonged, or if a trace of hydrochloric acid 
 is present. Hence any residue remaining after the treat- 
 ment with nitric acid, unless it is evidently sulphur, should 
 be examined. 
 
 The residue insoluble in nitric acid which was left in the 
 first dish, is to be washed with water in order to remove 
 the adhering solution. This may generally be done by pour- 
 ing water into the dish, allowing the precipitate to settle 
 and then decanting the wash- water; it is sometimes neces- 
 sary, however, to collect the precipitate on a filter and wash 
 in the ordinary manner. In either case the wash-water is 
 thrown away and the precipitate is boiled in the porcelain 
 dish with as much aqua regia (App., § 7) as will barely 
 cover it. Dilute the acid solution obtained with an equal 
 volume of water, remove from it, by filtration or other- 
 wise, any particles of free sulphur which may remain 
 undissolved, and add to it almost, but not quite, enough 
 ammonia- water to neutralize its acidity. In case of the 
 accidental addition of too much ammonia-water, manifested 
 by the appearance of a precipitate and by the alkaline 
 reaction on litmus, the solution must be made just acid by 
 the cautious addition of nitric acid, a drop at a time. Place 
 in the slightly acid solution a small bit of bright -pest 
 copper wire, and observe that metallic mercury is for 
 deposited upon the copper as a white silvery ^S- 
 coating. After the lapse of ten or fifteen minutes, dry the 
 wire upon blotting paper, drop it into a narrow glass tube 
 
26 SEPARATION OF BISMUTH. [§ 21. 
 
 which has been sealed at one end, and heat it at the lamp. 
 Metallic mercury will sublime, and be deposited as a dull 
 mirror upon the cold portions of the glass. By scratching 
 the sublimate with the point of a bit of iron wire, the metal 
 may be made to collect into visible globules. 
 
 When the greater part of the free nitric acid has, by the 
 
 aforesaid evaporation, been driven off from the filtrate 
 
 which contains the mixed nitrates of lead, bismuth, cadmium 
 
 Test ^i^cl copper, transfer the residual liquor to a test- 
 
 for tube, mix it with two or three times its volume of 
 
 ^^- dilute sulphuric acid, add half its own volume of 
 
 alcohol, and leave the mixture at rest during fifteen or 
 
 twenty minutes. Sulphate of lead will be thrown down as 
 
 a white powder, plainly to be seen in the test-tube, though 
 
 it would have been scarcely visible in the white dish. 
 
 Collect the filtrate from the sulphate of lead in a small 
 beaker, and add to it ammonia-water by repeated small 
 portions, 'taking care to stir the liquid thoroughly after 
 each addition of the ammonia, until a strong, persistent 
 odor of the latter is perceptible. The hydrates of copper, 
 cadmium and bismuth will all be thrown down at first, 
 but the hydrates of copper and cadmium will redissolve in 
 the excess of ammonia-water, and hydrate of bismuth will 
 alone be left as an insoluble precipitate. The blue color of 
 the solution is due to the presence of copper. If in the case 
 of the actual examination of a solution of unknown com- 
 position no precipitate falls on the addition of the ammo- 
 nia-water, this does not prove the absence of copper and 
 cadmium, as the hydrates may be dissolved by the ammo- 
 nia-water as fast as formed, and thus escape observation. 
 
 To prove that the precipitate contains bismuth : — Collect 
 
 Test it upon a filter, allow it to drain, and dissolve it 
 
 for in the smallest possible quantity of strong hydro- 
 
 ^*- chloric acid poured drop by drop upon the sides 
 
 of the filter; carefully evaporate the acid solution to the 
 
§ 21.] SEPARATION OF CADMIUM. 27 
 
 bulk of two or three drops, and pour it into a large test- 
 tube nearly full of water. A dense, milky cloud of insol- 
 uble basic chloride of bismuth will appear in the water. 
 Since sulphate of lead is not absolutely insoluble in water 
 which contains nitric acid, a slight precipitate of hydrate 
 of lead might be produced on the addition of the ammonia- 
 water even when no bismuth was present in the solution. 
 To prove the presence of bismuth, the oxychloride must 
 always" be <?arefully tested for. It is to be remarked that if 
 the amount of hydrate of bismuth is in any case small, con- 
 siderable jiare must be exercised in applying the confirma- 
 tory test. In such a case the basic chloride may appear 
 as white threads marking the path of the heavier liquid 
 through the water, and gradually producing more or less 
 turbidity throughout the whole mass of the water. 
 
 The blue color of the ammoniacal filtrate from the hydrate 
 of bismuth indicates the presence of copper, and when well 
 defined is of itself a sufficient proof of the presence of 
 this element. But in the absence of a marked blue colora- 
 tion, at this stage, copper should be specially tested for in 
 the manner described below. 
 
 To prove the presence of copper and cadmium, proceed as 
 follows : — Divide the ammoniacal filtrate into two portions. 
 
 To prove the presence of copper in case no blue colora- 
 tion has appeared : — Concentrate one portion in an evapo- 
 rating-dish, acidify with acetic acid (App., § 13), Test 
 transfer to a test-tube, and add one or two drops of for 
 a solution of ferrocyanide of potassium (App., § 34). ^^• 
 A peculiar reddish-brown precipitate of ferrocyanide of cop- 
 per will fall in case much copper is present, and even when 
 the proportion of copper in the solution is extremely small 
 a light brownish-red cloudiness will be produced. 
 
 To prove the presence of cadmium : — Add cyanide of 
 potassium solution (App., § 36) to the other portion of 
 the alM.ine filtrate (if blue coloration showing the pres- 
 
28 SEPARATION OF CADMIUM. [§§^1,22. 
 
 ence of copper has been visible, the whole of the alkaline 
 filtrate may be used) until decolorized; if no blue color 
 has appeared, only a few drops of the cyanide of potassium 
 Test solution need be added; pass sulphuretted hydro- 
 for gen through the solution, or add sulphuretted 
 ^^- hydrogen water. The liquid immediately be- 
 comes cloudy from the presence of minute particles of 
 sulphide of cadmium of characteristic yellow color. In 
 case an exceedingly small amount of cadmium iT'present 
 there may be only a yellow coloration: the sulphide can 
 even in this case be rendered visible by fillRffion, the 
 yellow precipitate remaining on the filter. If on passing 
 sulphuretted hydrogen through the solution a black or dark- 
 colored precipitate is obtained, it may be due to the pres- 
 ence of lead or mercury — in the former case because the 
 lead was not completely removed as sulphate, in the latter 
 case because the element was not completely converted into 
 the black sulphide. In case other than a yellow precipitate 
 appears, filter and wash the precipitate, boil it in an evapo- 
 rating-dish with dilute sulphuric acid (App., § 11), filter 
 again and pass sulphuretted hydrogen through the filtrate; 
 but if the precipitate is still black, owing to the presence of 
 sulphide of mercury, it must be boiled with concentrated 
 hydrochloric acid, which will dissolve the sulphide of 
 cadmium if present: the solution is then largely diluted 
 with water and again tested with sulphuretted hydrogen. 
 
 22. The operations above described may be presented in 
 tabular form as follows : — 
 
f§ 22, 23.] SEPARATION OF CLASSES I AND II. 
 
 29 
 
 The General Keagent (H^S) of Class II precipitates HgS, PbS, 
 Bi^Sa, CdS and CuS (as well as members of Class III, which 
 are subsequently separated by solution in sulphide of sodium). 
 The precipitate is boiled with dilute nitric acid : — 
 
 A residue of 
 HgS, mixed 
 with S, re- 
 mains. 
 
 Confirm 
 presence of 
 Hg with cop- 
 per w^e. 
 
 ,Pb, Bi, Cd and Cu go into solution as nitrates. 
 On addhig dilute sulphuric acid and half its volume 
 of alcohol to the concentrated solution : — 
 
 PbSO^ i s 
 thrown down. 
 
 The sulphates of Bi, Cd and Cu 
 
 remain in solution. On adding an ex- 
 cess of ammonia- water : — 
 
 Hydrate of 
 Bismuth is 
 thrown down. 
 Confirm Bi 
 by precipitat- 
 ing the oxy- 
 chloride. 
 
 ' Compounds of Cd 
 and of Cu remain in 
 solution. 
 
 Divide into two por- 
 tions. A and B. 
 
 A. Acid- 
 ify with 
 acetic acid. 
 Confirm 
 presence 
 of copper 
 by testing 
 with f erro- 
 cyanide of 
 potassium. 
 
 B. Add 
 
 cyanide of 
 potassium. 
 Confirm 
 presence 
 of cadmi- 
 um by pre- 
 cipitation 
 of CdS. 
 
 23. The Method of Separating Class I from Class II has 
 already been x^articularly described in § 6. It should be 
 observed, however, that even if no member of Class I were 
 present in the mixture to be analyzed, it would still be 
 necessary to acidulate the liquid with hydrochloric acid, 
 before passing the sulphuretted hydrogen, in order to 
 prevent the precipitation of members of Classes IV and V, 
 and to secure the complete precipitation of members of 
 Class III. 
 
 The liquid should be watched attentively when the 
 stream of sulphuretted hydrogen first begins to flow through 
 it, since useful inferences may often be drawn from the 
 various phenomena which present themselves. 
 
30 SEPARATION OF CLASSES II AND III, [§ 23. 
 
 a. Tims, the formation of a white precipitate which 
 afterwards changes to yellow, orange, brownish-red, and 
 finally to black, as the liquid gradually becomes saturated 
 with the gas, indicates the presence of mercuric chloride. 
 In order to exhibit this reaction, pass sulphuretted hydrogen 
 through two or three teaspoonfuls of a solution of mercuric 
 chloride. The white precipitate at first formed is a com- 
 pound of chloride and sulphide of mercury, but by the action 
 of successive portions of sulphuretted hydrogen the com- 
 position and appearance of the precipitate is changed, until 
 it has been completely converted into black sulphide of 
 mercury. It is important that the mercury should be com- 
 pletely converted into sulphide, as the compound just men- 
 tioned is soluble in nitric acid : the result is rendered more 
 certain if the solution be warmed before being treated with 
 sulphuretted hydrogen. 
 
 b. If the precipitate is of a dull red color at first, after- 
 wards changing to black, the probable presence of lead is 
 indicated; for sulphuretted hydrogen throws down from 
 solutions which contain much free hydrochloric acid a red 
 compound of chloride of lead and sulphide of lead, which is 
 afterwards decomposed, with formation of the black sul- 
 phide, when the solution becomes saturated with the gas. 
 
 c. A decided bright yellow precipitate would indicate the 
 presence of cadmium, arsenic or tin; of these three, cadmium 
 is distinguished by the fact that its sulphide remains 
 undissolved when the precipitate is treated with sulphide 
 of sodium to separate the members of Class III. 
 
 But even from solutions which contain no members of 
 Classes II or III, yellow-white or milky-white precipitates 
 of free sulphur are often thrown down; for sulphuretted 
 hydrogen is easily decomposed, with deposition of sulphur, 
 by a variety of oxidizing agents, such as nitric, chromic 
 and chloric acids, and solutions of ferric salts and of free 
 chlorine. If the solution under examination contained 
 much nitric acid, sulphuretted hydrogen would have to be 
 
§ 23.] SEPARATION OF CLASSES II AND III. 31 
 
 passed through it for a long time to destroy the acid, before 
 the liquid could be saturated with the gas. • In this case 
 the sulphur separates as a tenacious mass of dirty yellow 
 color; but in most instances, notably when the solution 
 contains a ferric salt, the sulphur is precipitated in the 
 form of exceedingly minute particles, which impart to the 
 solution a peculiar milkiness or opalescence. These parti- 
 cles are so fine that they pass through the pores of filter 
 paper ; they cannot be removed by filtration. To see this 
 separation of sulphur, pass sulphuretted hydrogen through 
 two or three teaspoonfuls of a solution of ferric chloride; 
 also through nitric acid diluted with an equal volume of 
 water. 
 
 If the original solution contains a chromate, its yellow or 
 reddish-yellow color will be changed to green by the action 
 of sulphuretted hydrogen; for the chromate is reduced to 
 the condition of chromic chloride : — 
 
 2 MjCrO^ + 10 HCl + 3 HjS = 2 CrClj + 3 S + 4 MCI + 8 HjO. 
 
 To show this reduction, pass sulphuretted hydrogen 
 through two or three teaspoonfuls of a solution of chromate 
 of potassium to which a few drops of hydrochloric acid have 
 been added. 
 
 The sulphur is set free in the form of the minute white 
 particles above described, and remains suspended in the 
 green liquid, looking not unlike a green precipitate. 
 
 d. The immediate formation of a black precipitate in- 
 dicates the presence of copper or bismuth, and it is to be 
 observed that either of these black precipitates would ob- 
 scure the colors of the other sulphides of the class, and 
 conceal them if present. 
 
 e. If no precipitate appears even when the liquid has 
 become saturated with the gas, or only a slight cloudiness 
 from the decomposition of the sulphuretted hydrogen, the 
 absence of every member of Classes II and III is, of course, 
 to be inferred. 
 
CHAPTER IV. 
 
 CLASS III. — ELEMENTS WHOSE SULPHIDES ARE INSOL- 
 UBLE IN WATER OR DILUTE ACIDS, BUT SOLUBLE IN 
 ALKALINE SOLUTIONS. 
 
 24. Example of the Precipitation of the Members of Class 
 III. — Place in a small beaker six or eight drops of a solu- 
 tion of arsenious oxid e in hydrochloric acid and similar 
 quantities of solutions of the chlorides of antimony and tin . 
 Pour in enough dilute hydrochloric acid to half fill the 
 beaker ; a white precipitate of the basic chloride of antimony 
 may form, which can be disregarded. 
 
 Pass sulphuretted hydrogen gas through the solution, in 
 the manner described in § 20, until the odor of the gas 
 persists. Then collect the precipitated sulphides upon a 
 filter, and rinse the precipitate once or twice with water. 
 
 In the analysis of any complex solution of unknown com- 
 position which might contain one or all of the members of 
 Class III, the sulphides of this class would, of course, all be 
 thrown down at the same time as those of Class II. (Com- 
 pare §§ 8, 22.) It will be well, therefore, for the sake of 
 illustration, for the student to dissolve the present precip- 
 itate in sulphide of sodium in order that he may begin the 
 treatment of Class III at the precise point at which this 
 class would be encountered in an actual analysis ; namely, 
 with the sulphides of the class in alkaline solution. To 
 this end, allow the washed precipitate to drain, spread out 
 the filter upon a plate of glass, scrape the precipitate from 
 the paper with a small spatula of platinum, horn or wood, 
 and transfer it to a porcelain dish. Pour lipou the precipi- 
 32 
 
§§ 24, 25.] SEPARATION OF CLASS III. 33 
 
 tate two or three times as much of a solution of sulphide of 
 sodium as would be sufficient to cover it, and boil the mix- 
 ture very cautiously, so as to avoid spattering. The pre- 
 cipitate will soon dissolve, and no solid matter will be left 
 suspended in the solution, excepting a few fibres of the 
 filter paper. If nearly complete solution does not ensue, 
 the undissolved portion is probably SnS; in which Q3.se 
 allow the undissolved substance to settle, decant off the 
 liquid and to the precipitate add a little of the yellow sul- 
 phide of sodium solution and warm, or a pinch of flowers 
 of sulphur may be added to the boiling sulphide solution. 
 When solution has taken place mix this liquid with that 
 just decanted and proceed. In the case of an actual analy- 
 sis when sulphides of members of Class II are present, the 
 yellow sulphide should always thus be finally employed 
 (or a little sulphur added as above), an excess being, how- 
 ever, avoided. It is such a solution as this which in an 
 actual analysis is examined for members of Class III by 
 either of the following methods. Divide the solution 
 obtained into two portions, using one portion for the first .^j ^ 
 method of analysis : the other for the second method. JjHB 
 
 25. Analysis of the Mixed Sulphides. — First methoo^^^ 
 This method depends, 1st. Upon the oxidation of the 
 several sulphides by means of nitric acid and nitrate of 
 sodium, and the conversion of arsenic into a compound 
 soluble in water, while antimony and tin are converted into 
 insoluble compounds ; 2d. Upon the fact that the tin in the 
 compound thus formed is set free in the metallic state when 
 this compound is treated with zinc and hydrochloric acid, 
 while the antimony under similar circumstances is in part 
 reduced to metallic antimony and in part conv^ed into 
 antimoniuretted hydrogen; 3d. Upon the solubility of tin 
 and insolubility of gold and platinum in strong hydrochloric 
 acid. 
 : To effect the -separation I: — Add strong nitric acid to the 
 
34 SEPABATION OF ARSENIC. [§ 25. 
 
 sulphide of sodium solution until the reaction is distinctly- 
 acid; a precipitate forms which consists partly of the sul- 
 phides of arsenic, antimony and tin, and partly of sulphur 
 from the decomposition of the alkaline sulphide. Without 
 heeding the precipitate which separates, evaporate the acid 
 mixture to the bulk of half a teaspoonfulj then add solid 
 carbonate of sodium ve^y cautiously until the last addition 
 causes no further effervescence. Then add about half a 
 teaspoonful of a mixture of equal parts of solid carbonate 
 and nitrate of sodium. Heat the mass with constant stir- 
 ring until it fuses. The fused mass should be white (if no 
 gold or platinum are present as in this case) and liquid. It 
 is sometimes necessary to add more of the mixture of nitrate 
 and carbonate of sodium towards the latter part of the 
 fusion, especially if too great an excess of the alkaline 
 sulphide has been used in making the solution. Pour out 
 the liquid mass, as far as may be possible, upon a bit of 
 cold porcelain, and when it has cooled somewhat, transfer 
 it, together with whatever can be detached from th% dish in 
 which the fusion was made, to a clean mortar and reduce it 
 to powder. Eeturn this powder to the porcelain dish, and 
 when the dish with its contents is perfectly cold pour in 
 several teaspoonfuls of cold water and allow the mixture 
 to stand, stirring it from time to time, until that portion of 
 the fused mass which could not be detached from the dish 
 has softened and is in part dissolved; finally, filter the 
 solution. By the treatment with nitric acid and nitrate of 
 sodium the arsenic of the sulphide of arsenic has been con- 
 verted into arseniate of sodium, a compound which is solu- 
 ble in water and which passes into the filtrate together with 
 the excess of nitrate and carbonate of sodium and some 
 sulphate of sodium formed during the process. The anti- 
 mony has been converted into antimoniate of sodium and 
 the tin into the binoxide; these two compounds, being 
 scarcely at all soluble in water, remain on the filter. (Gold 
 
§ 26.] SEPARATION OF ARSENIC. 35 
 
 and platinum if present in the original solution will here 
 be in the metallic condition.) This insoluble residue is 
 examined for antimony and tin in a manner presently to be 
 described. 
 
 To the filtrate add nitric acid drop by drop until the 
 liquid shows a faintly acid reaction; warm gently to remove 
 all traces of carbonic acid set free by the action of the nitric 
 acid on the excess of carbonate of sodium used in the fusion. 
 If a precipitate appears on addition of nitric acid, it is 
 probably a hydrate of tin, and its appearance is due to the 
 fact that the fusion with nitrate of sodium was carried on 
 at so high a temperature that a portion of the nitrate of 
 sodium was converted into oxide of sodium, and this acting 
 on the oxide of tin formed a certain amount of soluble 
 stannate of sodium. The precipitate, if any appear, should 
 be filtered off and added to the insoluble residue xest 
 which awaits examination. To one half of the for 
 clear filtrate add ammonia-water to alkaline reac- "^^• 
 tion, then a few drops of a prepared solution of chloride of 
 magnesium and chloride of ammonium (App., § 48) and 
 set the mixture aside for twelve hours. The formation of 
 a white crystalline precipitate of arseniate of ammonium and 
 magnesium is evidence of the presence of arsenic. 
 
 To the remainder of the filtrate which contains the arsenic, 
 and which should contain the smallest possible quantity of 
 free nitric acid, add a small quantity of a solution of nitrate 
 of silver. On the addition of nitrate of silver a precipitate 
 generally falls even in the absence of arsenic, owing to the 
 fact that the nitrate of sodium used in the fusion usually 
 contains some chloride of sodium, and in order that the 
 arsenic present may be converted into arseniate of silver a 
 small quantity of nitrate of silver must be added over and 
 above that necessary to remove the chlorine present. The pre- 
 cipitated chloride of silver is removed by filtration. To 
 the clear filtrate add a solution of pure acetate of sodium 
 
36 SEPARATION OF ANTIMONY. [§25. 
 
 (App., § 28) drop by drop, mixing thoroughly after each ad- 
 dition, until the mixture smells of acetic acid; a red or 
 brownish-red precipitate of arseniate of silver appears. 
 The acetate of sodium is added because the arseniate of 
 silver is soluble in nitric acid, and but sparingly soluble in 
 acetic acid unless it is present in considerable excess. On 
 the addition of acetate of sodium to the liquid containing 
 free nitric acid there was formed nitrate of sodium, and 
 acetic acid was set free. That the test be satisfactory, the 
 following precautions must be carefully observed. The 
 filtrate from the fusion must be distinctly acid with nitric 
 acid and free from all traces of carbonic acid, but must not 
 contain a great excess of the acid. Care must be taken that 
 enough nitrate of silver is added and that the solution of 
 acetate of sodium contains no organic matter, which often 
 appears in solutions of this reagent which have been kept 
 for some time. 
 
 Any precipitate supposed to contain arsenic (if in the 
 form of sulphide, it must be converted into some other com- 
 pound) may be further examined by converting the arsenic 
 into a hydrogen compound by the method known as " Marsh's 
 test," which consists in converting the arsenic into arseniu- 
 retted hydrogen in a similar manner to that employed in 
 the case of antimony about to be described. For fuller 
 details of "Marshes test" the student is referred to any 
 standard work on General Chemistry. This method is 
 applicable to the detection of very minute quantities of arse- 
 nic, and owing to the delicacy of the test, if arsenic is present 
 in the original solution some trace of it will usually appear 
 in the test for antimony here described. 
 
 The insoluble residue containing antimony and tin, which 
 was left on the filter, is washed several times with a mix- 
 ture of equal parts of alcohol and water, and then trans- 
 ferred to a test-tube or porcelain dish and warmed with 
 strong hydrochloric acid. Choose a cork or caoutchouc 
 
§25.] SEPARATION OF ANTIMONY. 37 
 
 stopper provided with two holes, which fits accurately the 
 mouth of a wide test-tube. Fit a small thistle-tube to one 
 of the holes, and to the other a short bent tube drawn to a 
 rather fine open point. Put a fragment or small strip of 
 zinc (App., § 57), together with a bit of platinum foil into 
 the tube, cover with water, close the tube with the perforated 
 stopper, and through the thistle-tube add strong hydrochlo- 
 ric acid in small successive portions. After hydrogen has 
 been generated freely during four or five minutes by the 
 mixture in the tube, and all the air originally contained in 
 the latter has been expelled, light the gas issuing from the 
 pointed glass tube, and pour into the thistle-tube, a portion 
 at a time, the mixture of strong hydrochloric acid and the 
 residue containing antimony and tin, or the solution if the 
 acid has dissolved the residue completely. Hold a cold 
 porcelain dish or bit of broken porcelain in the flame, taking 
 care to shift the position of the dish frequently so that fresh 
 surfaces of porcelain may be exposed to the burning gas. 
 If there be really any antimony in the insoluble residue, an- 
 timoniuretted hydrogen will be evolved, together with free 
 hydrogen, and characteristic smoky-black spots or stains of 
 metallic antimony will be deposited from it upon xest 
 the cold porcelain. To be sure that the spots are for 
 really composed of antimony and not of arsenic, ^*'- 
 which forms arseniuretted hydrogen and deposits metallic 
 arsenic under the circumstances in which antimoniuretted 
 hydrogen is produced and deposits antimony, cover them 
 with a solution of hypochlorite of sodium (App., § 63) : if 
 they are antimony spots, they will not dissolve, being unaf- 
 fected for a long time, while arsenic spots dissolve at once. 
 The following comparisons of arsenic and antimony thus 
 deposited may be noticed. The arsenic spots have a 
 metallic lustre and brown color when thin; the stain of 
 antimony has a feeble lustre and is smoky black. The 
 arsenic spot disappears readily at a heat below redness; the 
 
S8 SEPARATION OF ANTIMONY AND TIN, [§ 25. 
 
 stain of antimony is volatile only at a red heat. The arsenic 
 spot is not perceptibly ali'ected by yellow sulphide of am- 
 monium unless heated : the antimony stain dissolves readily, 
 a bright orange stain remaining after evaporation. 
 
 In order not to explode the test-tube on lighting the gas, 
 the operator must wait patiently for several minutes, until 
 all the air has been expelled from the tube. 
 
 The whole of the antimony is not converted by this means 
 into antimoniuretted hydrogen ; a portion is reduced to the 
 metallic state, and if the evolution has not been too rapid, 
 will be found deposited on the platinum foil as a firmly ad- 
 herent dark coating or stain. Therefore when the zinc in 
 the test-tube is nearly all consumed, transfer the contents 
 of the tube to a porcelain dish and examine the platinum 
 foil for this indication of the presence of antimony. 
 
 To confirm the presence of antimony indicated by the 
 black stain or coating on the platinum foil, warm the foil 
 for a few moments with strong hydrochloric acid to remove 
 all traces of tin, wash free from the acid, treat with a few 
 drops of dilute nitric acid and a few drops of tartaric acid 
 (App., § 14), heat nearly to boiling, dilute with an equal 
 volume of water and pass sulphuretted hydrogen through 
 the solution. An orange precipitate shows the presence of 
 antimony. 
 
 In the operation of testing for antimoniuretted hydrogen 
 as above described, the tin which was in the form of oxide 
 was reduced to the metallic state and now awaits examina- 
 tion in the porcelain dish. Remove and rinse any zinc 
 remaining undissolved and carefully decant from the dark 
 
 Test spongy mass the solution with which it is covered, 
 for and which consists mainly;? of chloride of zinc. 
 
 S"- The residue is warmed with strong hydrochloric 
 acid, in which the tin dissolves. 
 
 Pour off the solution of stannous chloride thus obtained, 
 and add to it a few drops of a solution of mercuric chloride 
 (App., ^55): a white or gray precipitate of mercurous chlo- 
 
§§ 26, 26.] 
 
 S^PAttATlON OF TIN. 
 
 89 
 
 ride often mixed with gray metallic mercury will be thrown 
 down; for — 
 
 2 HgCl^ + SnClj = 2 HgCl + SnCl,, and 2 PIgCl + SnCl^ - 2 Hg + SnCl,. 
 
 To prove that the precipitate really contains mercurous 
 chloride, decant the supernatant liquid; cover the precipi- 
 tate with ammonia-water and heat the mixture to boiling 
 (compare j). 21). A black residue slowly soluble in hot 
 concentrated hydrochloric acid is often obtained when no 
 tin is present, which is metallic lead, a common impurity in 
 zinc. Enough of this lead may dissolve in the strong hydro- 
 chloric acid to give a white precipitate of chloride of lead 
 when the liquid is diluted by the addition of the solution of 
 mercuric chloride. Hence it is always best, in case a white 
 precipitate is formed, to prove that it is really mercurous 
 chloride by treating with ammonia-water as described. 
 
 A black residue insoluble in strong hydrochloric acid 
 may be metallic gold or platinum if either of these elements 
 were present in the original solution, or perhaps it may be 
 metallic lead from the impure zinc. 
 
 26. An outline of the foregoing operations may be repre- 
 s^ted in tabular form as follows : — 
 
 The General Reao;ent (H.^S) of Class III precipitates As-^S.^, 
 Sb.^S^ and SnS or SnS,„ [An^^] and [EtB^]. (As well as the 
 members of Class II from which Class III is separated by solution 
 in sulphide of sodium.) The sulphide of sodium solution is treated 
 with nitric acid and carbonate of sodium; the mixture is evaporated, 
 mixed with carbonate and nitrate of sodium and heated to fusion. 
 The fused mass is treated with cold water. 
 
 Arseniate of sodium 
 (with nitrate of sodium, 
 etc.) goes into solution. 
 Confirm presence of As 
 by magnesium mixture 
 and by the silver test. 
 
 Antimoniate of sodium and oxide of tin 
 remain undissolved. Reduce with zinc 
 and HCl in presence of platinum foil. 
 
 A ntimoniuretted 
 hydrogen is formed 
 and antimony spots 
 obtained. 
 
 Tin is left in the 
 metallic state (to- 
 gether with some an- 
 timony which stains 
 the foil). Dissolve 
 the tin in HCl and 
 test with HgCl.,, 
 
40 SEPARATION OF GOLD AND PLATINUM. [§ 26. 
 
 In the case of mixtures containing gold and platinum (see 
 § 13), as well as arsenic, antimony and tin, the gold and 
 platinum would remain with the tin, without interfering in 
 any way with the separation or detection of either member 
 of the class, and when the tin reduced by zinc is dissolved 
 in hydrochloric acid, the gold and platinum would remain 
 undissolved (together with some antimony, if present). 
 Since the sulphides of gold and platinum are both black, 
 while those of arsenic and tin are yellow or brown, and that 
 of antimony is orange, the presence of any considerable 
 quantity of either of the precious metals would be indicated 
 by the black color of the class precipitate. 
 
 There are excellent special tests both for gold and for 
 platinum, by which these elements may be detected even in 
 the presence of all the other metals. Hence it is most con- 
 venient to make special search for them in the original 
 substance, by methods to be described hereafter (§ 98, 6), 
 whenever the preliminary examination has given reason to 
 suspect the presence of either of them. 
 
 In the regular course of analysis these elements may be 
 detected by the following method. 
 
 Wash the black residue insoluble in concentrated hydro- 
 chloric acid thoroughly free from acid; add to it a few 
 drops of nitric acid and of tartaric acid; heat gently to 
 remove any antimony that may be present in metallic con- 
 dition; wash thoroughly; heat the black residue in an 
 evaporating-dish with a little aqua regia; evaporate to a 
 small volume; add a few drops of chloride of ammonium; 
 
 Test evaporate nearly to dryness at a gentle heat, and 
 for treat the residue with a teaspoonful or so of a mix- 
 ^*- ture of equal parts alcohol and water. A yellow 
 residue appearing crystalline under the microscope indi- 
 cates platinum. 
 
 Eemove the platinum, which is in the form of chloro- 
 platinate of ammonium, by filtering through a small filter, 
 
§§ 26,27.] ALTEttNATtVM METHOD FOU CLASS III. 41 
 
 warm gently to remove the alcohol, and add a few drops of 
 oxalate of ammonium solution and of oxalic acid; ^pest 
 a brown or violet precipitate often forming slowly for 
 indicates gold, separating in the metallic condition. '^^• 
 
 27. As considerable difficulty is often experienced, 
 especially by beginners, in analyzing the mixed sulphides 
 of this class, the following method of treatment is suggested 
 as an alternative. 
 
 To effect the separation : — Add dilute hydrochloric acid 
 in small portions to the sulphide of sodium solution until 
 the reaction is distinctly acid. The sulphides of arsenic, 
 antimony and tin are precipitated, together with some 
 free sulphur. Any large excess of acid is to be avoided 
 lest the sulphides of antimony and tin, which are some- 
 what soluble in hydrochloric acid, fail to be precipitated. 
 
 Collect the precipitate on a filter and throw the filtrate 
 away. The precipitate is rinsed with water, transferred 
 to a test-tube or small flask, covered with strong hydro- 
 chloric acid, and gently warmed, not boiled. The yellow 
 residue is sulphide of arsenic. The antimony and tin pass 
 into the filtrate as chlorides. To test for these elements, 
 prepare a small hydrogen-generator, as described at the 
 top of page 37, and test for the elements as there directed, 
 taking care to warm the hydrochloric acid solution to drive 
 off all traces of sulphuretted hydrogen before placing it in 
 the hydrogen-generator. 
 
 To test for arsenic, transfer the sulphide of arsenic which 
 did not dissolve in the strong hydrochloric acid to a small 
 flask, add a teaspoonful of strong nitric acid, and warm 
 gmtly for several minutes. Then add in small successive 
 ])ortions chlorate of potassium, using in all an amount no 
 larger than half a pea. 
 
 When the sulphide of arsenic has entirely dissolved, 
 pour the liquid into an evaporating-dish and warm gently 
 until the chlorous smell has disappeared. This treatment 
 
42 SEPARATION OF CLASSES It AND lit. [§§ 27, 28. 
 
 for arsenic must be carried on where there is a good draught, 
 and the acid must be warmed, not boiled. When the chlo- 
 rous fumes have been driven off, add ammonia-water until 
 the reaction is alkaline, filter, if necessary, and to the clear 
 alkaline liquid add a small quantity of a prepared solution 
 of chloride of magnesium and chloride of ammonium (App., 
 § 48), and set the mixture aside for twelve hours. The 
 formation of a white crystalline precipitate of arseniate of 
 ammonium and magnesium is evidence of the presence of 
 arsenic. 
 
 In this method, if gold and platinum be present, they 
 will be found in the solution containing the arsenic; and if 
 a black residue remains after warming with strong hydro- 
 chloric acid, dissolving on addition of chlorate of potassium 
 and nitric acid, a portion of the solution should be tested 
 for these elements. 
 
 To test for gold and platinum, evaporate a portion of the 
 solution obtained as described almost to dryness, prepare a 
 small hydrogen-generator as in testing for antimony and 
 tin, and add this solution, with some strong hydrochloric 
 acid, which may contain arsenic, gold and platinum, just 
 as the solution of antimony and tin was added. Test the 
 ignited gas for arsenic as described in the test for antimony. 
 Gold and platinum will remain in the generator and may be 
 tested for as previously described. 
 
 28. The Method of Separating Class II from Class HI has 
 been sufficiently described in §§ 8, 24. When members of 
 Class II are altogether absent, something may be learned 
 from the color of the precipitate produced by sulphuretted 
 hydrogen. Thus, — 
 
 An orange-colored precipitate indicates the presence of 
 antimony; 
 
 A bright yellow precipitate, the presence of arsenic; 
 
 A dull yellow precipitate, white at first, the presence of 
 stannic sulphide; 
 
§ 28.] SEPARATION OF CLASSES II AND III 43 
 
 A dark brown precipitate, the presence of stannous sul- 
 phide ; 
 
 A black precipitate, the presence of gold or platinum. 
 
 When, as a result of the preliminary examination (§ 82, 
 III, cl), there is reason to suspect the presence of mercury 
 as a mercuric salt, sulphide of ammonium which has become 
 yellow from standing, should be substituted for sulphide of 
 sodium, because sulphide of mercury is somewhat soluble 
 in sulphide of sodium This may be seen by taking a few 
 drops of mercuric chloride solution, adding an equal volume 
 of water and two or three teaspoonfuls of sodium sulphide 
 solution. Boil, and filter if necessary. To a portion of the 
 clear filtrate add dilute nitric acid and observe the color of 
 the precipitate. Repeat the experiment, using sulphide of 
 ammonium instead of sulphide of sodium. In the event 
 of the analysis of an unknown solution, if on the treat- 
 ment with nitric acid of the solution of sulphides in sul- 
 phide of sodium, a black precipitate appear, it may be owing 
 to the presence of sulphide of mercury, and not to gold and 
 platinum; it is therefore sometimes better in such a case to 
 dilute the solution with water, filter and treat the precipi- 
 tate with sulphide of ammonium. The precipitate remain- 
 ing undissolved is added to the regularly obtained precipi- 
 tate of Class II, and the sulphide of ammonium solution 
 added, with a fresh portion of strong acid, to the nitric acid 
 solution already obtained. 
 
 If the second method has been employed, any sulphide of 
 mercury dissolved by the sulphide of sodium will remain 
 with the sulphide of arsenic. In case, therefore, this resi- 
 due is black or dark colored, it may be owing to the presence 
 of mercury, and not to gold or platinum. For this reason, 
 it may be advisable in such a case to test the dark-colored 
 residue with sulphide of ammonium. The residue remain- 
 ing undissolved is added to the regular Class II precipitate, 
 the sulphide of arsenic is reprecipitated by the addition of 
 
44 SEPARATION OF CLASSES II AND III [§ 28. 
 
 hydrochloric acid to the alkaline solution, and the precipi- 
 tate tested for arsenic as described. 
 
 If an abundant supply of the substance under examina- . 
 tion be at hand, it will often be better to start with a fresh 
 portion and make the separation of the two classes with 
 sulphide of ammonium. 
 
 The separation of the members of Class III from the 
 other elements precipitated by sulphuretted hydrogen in 
 acid solutions depends upon the formation of sulphur salts, 
 of arsenic, antimony, etc., when their sulphides are treated 
 with alkaline sulphides, which are in most cases readily 
 soluble. 
 
 In the course of a general analysis of a solution contain- 
 ing members of both the second and third classes, it is 
 essential that the precipitate produced by sulphuretted hy- 
 drogen should be washed free from acid to prevent decom- 
 position of the alkaline sulphides and introduction of 
 complications thereby. Sulphide of copper is somewhat 
 soluble in solutions of the normal sulphides of ammonium 
 and sodium, acids reprecipitating it sometimes as a light- 
 colored sulphide, which, with excess of sulphur, has nearly 
 the color of stannous sulphide or of arsenious sulphide. 
 Some loss may thus result, but enough of the sulphide, 
 which is entirely insoluble in the yellow sulphide of sodium 
 (sodium polysulphide, Na.^S^), will usually remain undis- 
 solved to insure the detection of copper in Class II. As 
 with mercury, the preliminary examination (§ 81) is of value 
 in this case also. 
 
 The normal colorless sulphides of ammonium and sodium 
 usually fail to dissolve stannous sulphide, as an excess of 
 sulphur is necessary for the formation of the soluble sul- 
 phostannates. 
 
 SnS + NajSj = NajSnSj. 
 
 Sulphide of ammonium is in general not so fit a solvent 
 
§ 28.] SEPARATION OF CLASSES II AND TIL 45 
 
 for the sulphides of Class III as sulphide of sodium; and if 
 the presence of tin is suspected, it is necessary to use the 
 yellow or polysulphide, or else add a little sulphur as men- 
 tioned. In any case it is important to use no more of the 
 alkaline sulphide than is necessary, on account of the 
 amount of sulphur which will have to be oxidized subse- 
 quently with nitric acid and nitrate of sodium. 
 
CHAPTER V. 
 
 CLASS IV. — ELEMENTS WHOSE HYDRATES ARE INSOLU- 
 BLE IN WATER, AMMONIA-WATER AND SOLUTIONS OF 
 AMMONIUM SALTS. 
 
 29. The leading fact upon which the separation of 
 this class is based is the insolubility of the hydrates of 
 iron, aluminum and chromium in ammonia- water, even in 
 presence of solutions of ammonium salts. But these 
 three hydrates are not the only substances which are liable 
 to be precipitated in an actual analysis when ammonia- 
 water is added in excess to a solution previously acid. 
 There are a number of compounds, soluble in acids, but not 
 in water or in weak alkaline liquids, which are thrown 
 down without change when their acid solvent is destroyed. 
 
 It is clear that it is needless to provide in this place 
 against the presence of such salts of elements belonging to 
 Classes I, II and III. Those elements are already elimi- 
 nated when the fourth class is taken in hand. But if there 
 are any salts of elements belonging to the fourth and higher 
 classes which can only be kept in solution by a free acid, 
 they will be precipitated without change in consequence of 
 the neutralization of their solvent by the ammonia-water 
 added to precipitate the three hydrates above mentioned. 
 Such salts are the phosphates of several members of Classes 
 IV, VI and VII, besides a number of oxalates, borates, 
 silicates and fluorides which occur so seldom that they 
 need not be particularly considered in an elementary treatise. 
 Beside the phosphates, several chromites and aluminates of 
 members of Classes VI and VII are insoluble in ammonia- 
 
§§ 29-31.] PRECIPITATION OF CLASS IV. 47 
 
 water, and are often thrown down wholly or in part along 
 with the legitimate members of Class IV. Manganese also 
 (a member of Class V) is frequently precipitated in com- 
 bination with members of Class IV, in the form of chromite, 
 ferrite or aluminate of manganese. The general scheme 
 for the examination of Class IV necessarily provides for 
 the detection of all the members of the class in the possible 
 presence of these extraneous substances. 
 
 30. Example of the Precipitation with Ammonia-water. — 
 Pour into a beaker a small teaspoonful of aqueous solu- 
 tions (App., § Q>Q)) of sulphate of manganese, c ommon a lum 
 (sulphate of aluminum and ammonium), c hrome alum (sul- 
 phate of chromium and potassium) and chh)ride_of , iron 
 (ferrous chloride). As an example of the substances insol- 
 uble in water which might be present in an acid solution, 
 dissolve in a small amount of boiling hydrochloric acid, a 
 not very large quantity, say half a gramme, of bone-ash 
 (phosphate of calcium), and add the solution to those 
 already placed in the beaker. Fill the beaker about one 
 third full of water, heat the mixture to boiling, and add to 
 it two or three drops of strong nitric acid to convert the 
 iron into ferric salts. Boil the mixture for a minute or 
 two and then add, little by little, ammonia-water to the 
 boiling liquor until a distinct odor of ammonia is percepti- 
 ble after the mixture has been thoroughly stirred. 
 
 31. Analysis of the Mixed Precipitate. — The following 
 method of detecting iron, chromium, aluminum (and man- 
 ganese) in the mixed precipitate which may contain all of 
 them, together with phosphates and other compounds of 
 barium, strontium, calcium and magnesium, depends: — 
 1st. Upon the oxidation and conversion of the hydrates of 
 manganese and chromium into manganate and chromate of 
 sodium (or potassium) when fused with a mixture of car- 
 bonate of sodium and nitrate of potassium; while the 
 hydrate of aluminum under the same treatment is converted 
 
48 SEPARATION OF CLASS IV. [§ 31. 
 
 to a greater or less extent, into aluminate of sodium, and 
 the compounds of barium, strontium and calcium either 
 remain unchanged or are converted into carbonates. 2d. 
 Upon the solubility of the chromate and aluminate of 
 sodium (or potassium) in water and the insolubility of the 
 carbonates or other compounds of barium, strontium and 
 calcium which may be present in the fused mass. 
 3d. Upon the sparing solubility of chromate of lead in 
 acetic acid. 4th. Upon the peculiar green color of the 
 manganate of sodium (or potassium). 5th. Upon the fact 
 that Prussian blue is formed when a solution of ferro- 
 cyanide of potassium is added to the solution of a ferric 
 salt. 6th. Upon the sparing solubility of the oxalates of 
 barium, strontium and calcium in dilute acetic acid. 
 — The details of the treatment of the precipitate produced 
 by ammonia-water, are as follows : — Collect the precipitate 
 upon a filter, wash it two or three times with water, and 
 then dry it either on the filter, or by transferring it to a 
 piece of platinum foil or to a platinum crucible (App., § 84) 
 and heating over the lamp gently so as to avoid spattering. 
 The dry precipitate is mixed intimately (best by rubbing 
 in a mortar) with five or six times its bulk of a dry mixture 
 of equal parts of carbonate of sodium and nitrate of potas- 
 sium (App., § 37). The mixture is then fused thoroughly 
 by heating it over the lamp either on platinum foil or in 
 a platinum crucible. If the amount of the mixture be 
 small, a piece of foil answers very well; larger quantities 
 may be fused in successive portions on the foil, but a small 
 crucible is much more convenient. 
 
 If only a manganese compound and no chromium had been 
 fused on the foil, the cold mass would have exhibited the 
 peculiar bluish-green color of manganate of sodium (or 
 potassium) owing to the oxidation of a small portion of 
 the hydrate of manganese to manganic acid in the presence 
 of carbonate of sodium and nitrate of potassium. If only 
 
§ 31.] SEPABATION OF CHROMIUM. 49 
 
 chromium had been present, the bright yellow color of 
 chromate of sodium (or potassium) would have been clearly 
 perceived. But from mixtures of the manganate and chro- 
 mate of sodium (or potassium), in various proportions, dif- 
 ferent shades of green, brownish-green or yellowish-green, 
 will result. When iron is present, the red color of its oxide 
 may obscure the colors due to manganese and chromium. . 
 
 Place the platinum foil or crucible in a porcelain dish, 
 cover it with water, and boil the latter until all the soluble 
 matter has been dissolved from the foil. Take out the 
 foil, rinse it, and throw the contents of the dish upon a 
 filter. The manganate, chromate and aluminate of sodium 
 pass into the filtrate, along with the excess of the carbonate 
 of sodium and nitrate of potassium employed : the filtrate 
 is colored yellow by the chromate. The insoluble residue, 
 in this case, consists of the oxides of iron and manganese, 
 together with the phosphate of calcium which has not been 
 altered by the fusion, and any small portion of carbonate 
 of calcium which may have been formed by the decomposi- 
 tion of a part of the phosphate. In the case of an actual 
 analysis there might be also phosphates, carbonates and 
 other insoluble compounds of barium, strontium and mag- 
 nesium, as well as of aluminum, iron, etc. 
 
 Divide the filtrate from the insoluble residue of the 
 fusion into two portions. Carefully add acetic acid, drop 
 by drop, to one of these portions until the liquor exhibits 
 an acid reaction, after all carbonic acid is set free, and then 
 add to it two or three drops of a solution of acetate of 
 lead (App., § 46). An insoluble precipitate of chromate of 
 lead will be immediately thrown down, exhibiting a bright 
 yellow color if the reagents be all pure. But if, as often 
 happens, the carbonate of sodium, employed as Test 
 the flux, is contaminated with sulphate of sodium, for 
 the yellow color of the precipitate will tend ^^• 
 towards white, in proportion to the amount of sulphate 
 
J 
 
 60 SEPARATION OF ALUMINUM. [§ 3L 
 
 of lead which has gone down together with the chromate. 
 A pure white precipitate would be no indication of 
 chromium, but only of a sulphate in the reagents. 
 
 Acidulate the other portions of the aqueous solution of 
 the fused sodium (and potassium) compounds with dilute 
 
 Test hydrochloric acid, add ammonia-water to slight 
 for alkaline reaction, warm the mixture and leave it 
 •^^- at rest for at least half an hour or, better, over 
 night. After the lapse of some time, a characteristic, gelat- 
 inous, colorless agglomeration of particles of hydrate of 
 aluminum will appear at the top or bottom of the liquid. 
 
 It should be said, that flocks of hydrate of aluminum, 
 when diffused through a liquid, are almost transparent 
 enough to elude observation. When an acid solution, con- 
 taining much aluminum, is mixed with ammonia-water and 
 warmed, a copious precipitate of hydrate of aluminum will 
 appear immediately, and will often remain floating for 
 some time upon the surface of the solution by virtue of 
 bubbles of air entangled in it. But since it is not easy to 
 convert the whole of the alumina in the original precipitate 
 into soluble aluminate of sodium, by fusion with carbonate 
 of sodium in the method above described, the quantity of 
 the hydrate to be thrown down at the final test is often 
 very small, and considerable time must be allowed, in 
 order that every particle of it may separate from the solu- 
 tion, and all the particles collect into a single mass. 
 
 To confirm the presence of aluminum, collect the hydrate 
 in the point of a small filter and allow it to drain. Cut 
 away the superfluous paper, place that portion of the filter 
 to which the precipitate is attached upon a piece of char- 
 coal, and heat it intensely in the blowpipe flame. Moisten 
 the residue with a drop of a solution of nitrate of cobalt 
 and again ignite it strongly. The unfused compound of 
 aluminum, cobalt and oxygen left upon the coal will exhibit 
 a deep sky-blue color when allowed to cool. This reaction 
 
§ 31.] SEPARATION OF MANGANESE AND IBON. 51 
 
 is useful in distinguishing the hydrate of aluminum from 
 that of glucinum, an element somewhat similar to alu- 
 minum though far less abundant. Hydrate of glucinum 
 when ignited with nitrate of cobalt does not yield a pure 
 blue compound, but only a gray mass. Silica gives a very 
 faint blue tint. 
 
 Eeturn now to the insoluble residue of the fusion with 
 carbonate of sodium and nitrate of potassium. If the char- 
 acteristic color of the manganate of sodium (or potassium) 
 were not distinctly observed at the previous fusion, take 
 a small quantity of the insoluble residue and fuse it with 
 twice its bulk of a mixture of carbonate of sodium and 
 nitrate of potassium upon platinum foil in a strong oxidizing 
 blowpipe flame (App., § 82). The peculiar bluish-green 
 coloration of manganate of sodium will appear ^^g^ 
 in the fused mass, as soon as it has become cold, for 
 particularly at the edges and thinner portions. In MSL- ~ 
 thus testing for manganese, it is well to incline the foil, so 
 that portions of the thoroughly melted mass may flow away 
 from the centre of the mixture into thin sheets, in order that 
 the color of the manganate may be exhibited in its purity. 
 
 Boil a second small portion of the insoluble residue with 
 a little strong hydrochloric acid, dilute the solution with 
 water, and add a drop or two of ferrocyanide of potassium. 
 The liquid will immediately become colored with xest 
 Prussian bine, an indication of the presence of for 
 iron. In case much iron be present, the blue color 1!®- 
 may be too deep to be recognized until the liquid has been 
 diluted with a large quantity of water. 
 
 Warm another portion of the insoluble residue with a 
 few drops of acetic acid, dilute the solution with water, 
 and filter if anything remain undissolved. To the slightly 
 acid liquid add a teaspoonful of a solution of oxalate of 
 ammonium (App., § 21). Any compounds of barium, stron- 
 tium and calcium which may have been present in the resi- 
 
52 SEPARATION OF CLASS IV. [§§ 31, 32. 
 
 due will be deposited as oxalates, as the oxalates of these 
 elements are but sparingly soluble in acetic acid. In the 
 prfesent case there will be a precipitate of oxalate of cal- 
 cium. Allow the mixture to stand for some time, and finally- 
 collect the precipitate on a filter, wash, dry and preserve 
 it for future examination in connection with Class VI. 
 
 In addition to the elements present in the solution which 
 has just been analyzed, compounds of magnesium are also 
 under certain circumstances precipitated with the hydrates 
 of Class IV. In actual practice, therefore, the remainder 
 of the insoluble residue is examined for magnesium, as fol- 
 lows : — Heat the mixture strongly on the platinum foil so 
 as to render the oxides of aluminum and iron as insoluble 
 as possible, and then treat with dilute hydrochloric acid 
 and warm gently. Without regarding the matter which 
 remains undissolved add to the mixture (which contains in 
 solution, along with the magnesium, some iron, aluminum, 
 calcium, etc.) a teaspoonful of chloride of ammonium, and 
 ammonia- water to alkaline reaction. Throw the mixture 
 upon a filter, collect the filtrate, and add it to that origi- 
 nally obtained from the precipitate of Class IV : this mixed 
 filtrate will be examined in due course (Class VII) for 
 magnesium. In case the whole amount of the residue 
 from the fusion of the precipitate of the class with car- 
 bonate of sodium and nitrate of potassium be but small, it 
 is well to filter the liquid of the preceding paragraph which 
 contains the precipitated oxalates of barium, strontium and 
 calcium, to evaporate the filtrate to dryness and ignite to 
 destroy the oxalic and acetic acids. The residue is subse- 
 quently treated with hydrochloric acid, chloride of am- 
 monium and ammonia, filtered and the filtrate added to the 
 filtrate from the original Class IV precipitate. 
 
 32. An outline of the foregoing operations may be tabu- 
 lated as follows : — 
 
§32.] 
 
 SEPARATION OF CLASS IV. 
 
 53 
 
 The General Reagent ([NHJHO mixed with NH^Cl) of Class 
 IV precipitates the hydrates of Fe, Cr and Al together with Mn 
 (as a chromite, ferrite or aluminate), and various phosphates and 
 other compounds of Fe, Cr, Al, Ba, Sr, Ca and Mg. The pre- 
 cipitate is dried and fused with Na-^COg and KNO3 ; the fused 
 mass is treated with water, and filtered : — 
 
 Divide the filtrate into two 
 portions : — 
 
 Divide the precipitate into four 
 portions : — 
 
 Yellow color 
 of the solution 
 indicates Cr. 
 Confirm Cr by 
 precipitation 
 of PbCrO,. 
 
 Acidulate 
 with HCl and 
 add (NH4) 
 HO. Color- 
 less flocculent 
 precipitate 
 proves pres- 
 ence of Al. 
 
 Test 
 for Mn 
 by fus- 
 ing with 
 Na^COs 
 
 and 
 KNO,. 
 
 Test for 
 Fe with 
 ferrocy- 
 anide of 
 potas- 
 sium. 
 
 Test for 
 Ca, etc., 
 by pre- 
 cipita- 
 tion of 
 the oxa- 
 lates 
 from 
 acetic 
 acid so- 
 lution. 
 
 Elimi- 
 nate Mg 
 by dis- 
 solving 
 in HCl 
 and re- 
 precipi- 
 tating 
 Class IV 
 with 
 NH^ 
 HO. 
 
 In the actual examination of an unknown substance, the 
 analysis is somewhat simplified, if the substance be a solid 
 soluble in water, or if it be a neutral solution, or if by any 
 other means we know that the phosphates, oxalates, etc., 
 mentioned above are absent. The treatment in such a case 
 may be simplified by the omission of those operations look- 
 ing to the separation of calcium, barium, etc., from the 
 residue of the fusion with carbonate of sodium and nitrate 
 of potassium. In the course of an analysis in case a white 
 precipitate falls on the addition of 'the general reagent of 
 Class ly, phosphoric and oxalic acids should be tested .for 
 at once (§§ 66 and 67). Phosphate of aluminum is not 
 easily decomposed by the fusion, nor is it soluble in acetic 
 acid, and it may remain as an insoluble residue after 
 treating the fused mass with water. In case, therefore, a 
 white residue remains insoluble in water and acetic acid, 
 it may be tested for aluminum by heating with nitrate of 
 cobalt on charcoal, as described in the confirmatory test for 
 
54 FERROUS AND FERRIC SALTS. [§§ 32, 33. 
 
 aluminum. Magnesium compounds give a flesh red or pink 
 colored mass by this treatment. 
 
 To determine whether the iron in the substance subjected 
 to analysis was originally in the state of a ferric or a fer- 
 rous salt, test a small quantity of the original solution 
 with a drop of ferricyanide of potassium (App., § 35). The 
 formation of Prussian blue proves the presence of a ferrous 
 salt. Another small portion of the original solution, tested 
 with a drop of ferrocyanide of potassium, would yield Prus- 
 sian blue in case the solution contained a ferric salt. In 
 applying either of these tests the blue coloration, indica- 
 tive of iron, is alone to be looked for; no notice need be 
 taken of other colorations, or of precipitates formed by the 
 action of the ferri- or ferro-cyanide upon the various metal- 
 lic salts which the solution may contain. The possibility 
 that a ferrous salt may have been changed into a ferric 
 during the process of getting the original substance, if a 
 solid, into solution, must not be lost sight of. 
 
 33. Separation of Class IV from Class III. The methods 
 of eliminating Classes I, II and III from mixtures which 
 contain members of these classes as well as of Class IV, 
 have already been described in §§ 6 and 8. 
 
 It is essential to the success of the operation that all 
 the sulphuretted hydrogen in the filtrate from Classes II 
 and III be expelled, for sulphuretted hydrogen precipitates 
 all the members of Classes IV and V from alkaline solu- 
 tions, and the filtrate now in question is, of course, made 
 alkaline when ammonia-water is added to it. The conver- 
 sion of the iron into a ferric salt (by means of nitric acid) 
 is necessary because ferrous hydrate is somewhat soluble 
 in ammonium salts, and could not, therefore, be precipi- 
 tated completely by ammonia-water in the acid filtrate from 
 Classes II and III. 
 
 No matter what the condition of the iron may have been 
 in the original solution, it is reduced to the state of fer- 
 
§ 33.] FERROUS AND FERRIC SALTS. 55 
 
 rous salts by sulphuretted hydrogen. The filtrate from the 
 precipitate produced by sulphuretted hydrogen (the general 
 reagent of Classes II and III) should, therefore, be placed 
 in a porcelain dish, and boiled, until the steam from it 
 ceases to blacken lead paper. After the sulphuretted 
 hydrogen has been expelled, three or four drops of strong 
 nitric acid must be added to the liquid, and the mixture 
 boiled for a moment longer to convert the iron into ferric 
 salts. If by accident the student should fail to convert the 
 iron entirely to the state of a ferric salt, there will be pro- 
 duced a greenish, slimy precipitate of ferrous hydrate on 
 the addition of the ammonia-water. This substance may 
 be seen by adding an excess of ammonia-water to a tea- 
 spoonful of ferrous chloride solution. Such a precipitate 
 must not be confounded with the green hydrate of chro- 
 mium : it should be redissolved in nitric acid and the acid 
 solution boiled anew for a few minutes before reprecipitat- 
 ing with ammonia-water. 
 
 When all the iron has been converted to the state of a 
 ferric salt, a small quantity of a solution of chloride of 
 ammonium is added to the boiling liquid, and finally am- 
 monia-water, little by little, with constant stirring, until 
 a persistent odor of ammonia is perceptible. A large excess 
 of ammonia must be carefully avoided, for hydrate of alumi- 
 num, being somewhat soluble in ammonia-water, might be 
 kept in solution, to the disturbance of the analysis of Classes 
 VI and VII. 
 
 Chromium, if present in the original substance in the 
 form of a chromate, must undergo reduction by the action 
 of sulphuretted hydrogen before it can be precipitated as a 
 hydrate in Class IV. In the course of a general analysis 
 no precipitate of chromium will be formed unless the ele- 
 ment is present as a salt of chromium, or has been reduced 
 to that condition by the action of the sulphuretted hydrogen 
 used to precipitate the members of Classes II and III. 
 
56 PBECIPITATES OF CLASS IV, [§ 33. 
 
 To illustrate this point, add ammonia-water to a solution 
 of chromate of potassium — acidulated with hydrochloric 
 acid, then pass sulphuretted hydrogen through an equal 
 amount of the chromate of potassium acidulated as before, 
 filter if necessary, boil the filtrate to drive off excess of 
 sulphuretted hydrogen and add ammonia-water in excess 
 as before. 
 
 It will be remembered that the object of using chloride 
 of ammonium is to hold in solution magnesium (of Class 
 VII) and the members of Class V. This it does in virtue 
 of the fact that the double salts formed by the union of 
 ammonium compounds with compounds of the elements in 
 question are soluble in water and also in ammonia-water. 
 A considerable quantity of the ammonium salt will, of 
 course, be formed in any event by the action of the ammonia- 
 water upon the hydrochloric acid in the solution, but it is 
 best always to add a further portion of the chloride as a 
 precautionary measure. 
 
 - The following inferences may be drawn from the color 
 of the precipitate produced by ammonia- water : — 
 
 A gelatinous, white precipitate indicates aluminum or 
 some one of the oxalates, phosphates, etc., mentioned above. 
 In this case test at once for phosphoric and oxalic acids. 
 
 A grayish-green or grayish-blue precipitate indicates 
 chromium. 
 
 A reddish-brown precipitate indicates iron. 
 
 If no precipitate is produced by the ammonia-water, all 
 the members of Class IV are absent, and the solution may 
 at once be tested with sulphide of ammonium, the general 
 reagent of Class V. 
 
 When the solution contains much chromium, a small por- 
 tion of this element is apt to remain dissolved at first in 
 the excess of ammonia-water, and to color the solution 
 pink; but by continuing to boil the solution, the color may 
 be made to disappear, and the whole of the chromium may 
 
§ 33.] ''MASHING'' OF CLASS IV. 57 
 
 be thrown down. Care must be taken to replace, by small 
 portions, the water driven off by boiling, lest some of the 
 members of Class Y be converted into insoluble compounds. 
 It is to be observed that the legitimate members of Class 
 IV cannot be completely precipitated by ammonia-water 
 from solutions which contain non-volatile organic substances, 
 like albumin, sugar, starch, and so forth, or organic acids 
 (such as tartaric, citric, oxalic, or even in some cases acetic 
 acid) which form soluble double salts by uniting simulta- 
 neously with the ammonium and one or more of the mem- 
 bers of the class. The treatment of substances containing 
 organic matter will be explained hereafter (§ 84). 
 
CHAPTER VI. 
 
 CLASS v. — ELEMENTS WHOSE SULPHIDES ARE INSOLU- 
 BLE IN WATER AND IN SALINE OR ALKALINE SOLU- 
 TIONS. 
 
 34. Example of the Precipitation of the Members of 
 
 Class V. — Place in a small glass flask a lialf teaspoonful of 
 aqueous solutions (App., § 66) of the sulphates, nitrates 
 or chlorides of cobalt, n ickel, m anganese a nd zinc. Add to 
 the mixture six or seven teaspoonfuls of a solution of 
 chloride of ammonium, as much water, and ammonia-water 
 to alkaline reaction. If any precipitate appear on the 
 addition of ammonia-water, add chloride of ammonium 
 solution and warm until it dissolves. 
 
 Heat the mixture to boiling, and add sulphide of am- 
 monium to the boiling solution, drop by drop, with fre- 
 quent agitation, as long as a precipitate continues to be 
 formed. (Compare p. 13.) In the present case there are 
 # special reasons why the precipitate should be boiled and 
 shaken, in order to make it compact ; for the sulphides of 
 Class V, when loose and flocculent, are not only easily acted 
 upon by the air and by dilute acids, but are peculiarly liable 
 to pass through the pores of filter paper, and yield muddy 
 filtrates. 
 
 At the best, these sulphides oxidize rapidly when moist, 
 with formation of soluble sulphates which are liable to pass 
 through the filters and contaminate the filtrates. The 
 analysis of the sulphides should therefore be proceeded 
 with immediately after the precipitation with sulphide of 
 
 58 
 
§§ 34, 35.] ANALYSIS OF THE MIXED SULPHIDES. 59 
 
 ammonium, and should be conducted in such manner that 
 no precipitate of a sulphide shall ever be left moist upon a 
 filter more than a quarter of an hour. 
 
 35. Analysis of the Mixed Sulphides. — The detection of 
 the several members of Class V depends : — 1st. Upon the 
 almost complete insolubility of the sulphides of cobalt and 
 nickel in cold dilute hydrochloric acid, and the ready solu- 
 bility of the sulphides of manganese and zinc in that liquid. 
 2d. Upon the solubility of hydrate of zinc, and the insolu- 
 bility of hydrate of manganese, in a solution of sodium 
 hydrate. 3d. Upon the insolubility of sulphide of zinc 
 in acetic acid, in presence of sulphuretted hydrogen. 
 4th. Upon the peculiar colors imparted to borax glass by 
 compounds of cobalt and nickel dissolved in the glass ; and 
 upon certain other special tests to be described directly. 
 
 To effect the separation : — Collect the precipitate upon a 
 filter, and rinse it once or twice with water ; spread open 
 the filter in a porcelain dish, and cover it with cold dilute 
 hydrochloric acid. Scarcely any of the sulphide of cobalt, 
 or of nickel, will go into 'solution, while the sulphides 
 of manganese and zinc will be completely decomposed, and 
 dissolved as chlorides. 
 
 Filter the hydrochloric acid solution, pour the filtrate 
 into a porcelain dish, and boil it until strips of moistened 
 lead paper held in the steam no longer indicate the pres- 
 ence of sulphuretted hydrogen; then add sodium hydrate 
 to the liquid in slight excess. A whitish gelatinous pre- 
 cipitate of hydrate of manganese, which turns brown on 
 exposure to the air and is insoluble in sodium hydrate, will 
 be thrown down, together with small portions of the 
 hydrates of cobalt and nickel, resulting from the partial 
 decomposition of the sulphides of these metals by the 
 hydrochloric acid, while the hydrate of zinc at first pre- 
 cipitated redissoives completely in the excess of the alka- 
 line hydrate. It is to be observed that precipitation should 
 
60 SEPARATION OF MANGANESE AND ZINC. [§ 35. 
 
 never be effected in a porcelain dish, since a white or 
 transparent precipitate is scarcely visible in a white and 
 opaque dish. 
 
 To prove the presence of manganese, collect the precipi- 
 Test tt'^te upon a filter, allow it to drain, and fuse a 
 for small portion of it with a mixture of carbonate of 
 ^"- sodium and nitrate of potassium upon platinum 
 foil in the oxidizing blowpipe flame, as directed on page 51. 
 Divide the alkaline filtrate into two portions, acidify one 
 portion with acetic acid, then pass sulphuretted hydrogen 
 Test through the liquid. Sulphide of zinc will be 
 for thrown down as a white or dirty white flocculent 
 ^^' precipitate. A slight decomposition of sulphu- 
 retted hydrogen whereby sulphur is set free, may give a 
 precipitate liable to be mistaken for sulphide of zinc; but 
 since the sulphide is soluble in hydrochloric acid, it may 
 be distinguished by adding a small quantity of this reagent 
 and noting whether most of the precipitate dissolves. Acid- 
 ify the other portion of the alkaline filtrate with hydro- 
 chloric acid. Add 07ie drop of a solution of nitrate of cobalt, 
 then a solution of carbonate of sodium as long as a precipi- 
 tate is produced. Boil for three minutes, filter, dry the 
 filter somewhat, and then burn either on platinum foil or 
 on charcoal. In the presence of zinc there remains an ash 
 of a peculiar bluish-green color. y^ 
 
 It is to be observed that in the analysis of mixtures 
 which contain no manganese the precipitate of hydrate of 
 cobalt or of nickel produced by the sodium hydrate is usu- 
 ally small and sometimes hardly perceptible ; but no matter 
 how minute the precipitate may be, it must always be care- 
 fully removed by filtration before testing the solution for 
 zinc with sulphuretted hydrogen, otherwise the white pre- 
 cipitate of sulphide of zinc will be obscured by the black 
 color of these sulphides. 
 
 The black residue, insoluble in dilute hydrochloric acid, 
 is washed with water and tested for cobalt and nickel, by 
 
§ 35.] SEPARATION OF COBALT AND NICKEL, 61 
 
 heating successive small portions of it in a bead (§ 91, e) 
 of borax (App., § '26) in tbe oxidizing blowpipe xests 
 flame. If cobalt alone were present, a bright, for 
 pure blue color would be imparted to the bead. Co&Ni. 
 Ofi the other hand, if the precipitate was composed solely 
 of sulphide of nickel, the borax glass would assume 
 a peculiar reddish-brown color. Mixtures of the two sul- 
 phides yield beads of various tints, according to the pro- 
 portions of nickel and cobalt contained in them. By 
 adding the precipitate to the borax by repeated small por- 
 tions, and fusing the bead anew after each addition, it is 
 often possible to obtain first the characteristic color of one 
 of the elements, and afterwards tolerably well defined indi- 
 cations of the other. 
 
 The blue color of cobalt can usually be made manifest, 
 even in presence of much nickel, by heating the borax bead 
 in the reducing blowpipe flame (App., § 82). In the re- 
 ducing flame the reddish-brown color imparted by nickel 
 changes to gray, while the cobalt blue remains unaltered. 
 
 In any event, one of the two metals will be detected by 
 the blowpipe test, and the subsequent operations can be 
 limited to searching for the other. 
 
 To prove the presence of nickel, boil the black residue 
 with a few drops of aqua regia in the porcelain dish, and 
 evaporate the solution almost, but not quite, to dryness. 
 Add to tJie residual acid liquor, little by little, a strong 
 solution of cyanide of potassium, until the reaction of the 
 solution becomes decidedly alkaline ; the cyanides of nickel 
 and cobalt at first thrown down both redissolve easily in 
 an excess of cyanide of potassium. Boil the mixture for 
 several minutes, adding water by small portions to replace 
 that lost by evaporation. Then add a solution of hypo- 
 chlorite of sodium until the liquid, after shaking, smells 
 strongly of it, and boil again. The nickel is precipitated 
 as black nickelic hydrate (NiH303), while the compound 
 of cobalt in solution (cobalticyanide of potassium) is not 
 
62 SEPABATION OF COBALT. [§§ 35, 36. 
 
 affected. If a light-colored precipitate appears, it may indi- 
 cate that an insufficient quantity of hypochlorite of sodium 
 has been added, and in the case of a solution actually under 
 examination more of the reagent should be added before 
 pronouncing nickel absent. The alkaline reaction of the 
 liquid must be maintained by the addition of more cyanide 
 of potassium, if necessary. 
 
 To confirm the presence of cobalt in case of doubt : — Dis- 
 solve the black residue in a few drops of hot aqua regia, 
 evaporate the solution nearly to dryness, pour into the 
 residual solution two or three times its own volume of a 
 solution of nitrite of potassium (App., § 38), and add to 
 the mixture concentrated acetic acid, until the reaction of 
 
 rpgg^ the liquid is strongly acid. Transfer the mixture 
 for to a test-tube, and leave it at rest during eighteen 
 
 ^°* or twenty -four hours. A beautif-ul, yellow crystal- 
 line precipitate of the double nitrite of cobalt and potassium 
 will be deposited sooner or later, according to the propor- 
 tion of cobalt which the solution contained. 
 
 On adding sodium hydrate to the filtrate from the cobalt 
 precipitate, hydrate of nickel would be thrown down if 
 present, and the presence of nickel might be confirmed by 
 testing this precipitate with borax in the oxidizing blow- 
 pipe flame. 
 
 36. An outline of the foregoing operations may be tabu- 
 lated as follows : — 
 
 The General Reagent [(NH4).^S] of Class V precipitates CoS, 
 NiS, MnS and ZnS. Treat the precipitate with dilute HCl : — 
 
 CoS and NiS 
 remain undis- 
 solved. 
 
 Test for Co 
 and Ni with 
 borax glass 
 and, if need be, 
 with NaClO or 
 KNO,. 
 
 MnCI, and ZnCl.^ go into solution. Boil, to 
 expel HJS, and add NaHO : — 
 
 Hydrate of manganese is 
 precipitated, together with 
 traces of the liydrates of 
 Co and Ni. 
 
 Prove presence of Mn by 
 the blowpipe test. 
 
 Hydrate of zinc 
 goes into solution. 
 Acidify with acetic 
 acid, and add H.^S 
 to throw down 
 ZnS. 
 
§ 37.] SEPARATION OF CLASSES IV AND V. 63 
 
 37. Separation of Class V from Class IV. After Classes 
 I, II, III and IV have been removed in the manner already 
 described (§§9, 33), add a single drop of sulphide of ammo- 
 nium of good quality (App., § 18) to the filtrate from Class 
 IV. If no precipitate is produced, none of the members of 
 Class V can be present, and the solution may be immedi- 
 ately tested with carbonate of ammonium, the general rea- 
 gent of Class VI. 
 
 If the first drop of the sulphide produces a precipitate, 
 transfer the mixture to a small flask, heat it until it actu- 
 ally boils, and add more of the sulphide, with the precau- 
 tions enjoined on page 13 to complete the precipitation. 
 
 In case the precipitate produced by sulphide of ammonium 
 is white, the presence of zinc is indicated. 
 
 If it be flesh-colored or yellowish-white and becomes 
 brown by oxidation when exposed to the air, the presence of 
 manganese is to be inferred. 
 
 The student may observe these changes by adding some 
 sulphide of ammonium to a solution of manganese chloride 
 or sulphate, collecting the precipitate on a filter and allow- 
 ing it to stand exposed to the air for an hour or two. 
 
 In case the precipitate is black, either cobalt or nickel, 
 or both these elements, are present. Both of them must 
 be sought for, whenever the precipitate exhibits any tinge 
 of black at the moment of its formation. 
 
 As the sulphide of nickel is not absolutely insoluble in 
 sulphide of ammonium, it not infrequently happens that 
 the filtrate from Class V is dark-colored, owing to the pres- 
 ence of sulphide of nickel in solution. (If it has a deep 
 brown color, nickel may be safely concluded to be present.) 
 In such case it is well to acidify the liquid with hydro- 
 chloric acid, and to collect the dark-colored precipitate 
 (which consists of sulphide of nickel mixed with free sul- 
 phur) on a filter, and to test for nickel by means of the 
 borax bead. The filtrate must then be made alkaline with 
 ammonia-water before testing for Class VI. 
 
64 SEPAttATION OF CLASSES IV AND V. [§ 37. 
 
 For'tlie sake of illustration the student may add an excess 
 of sulphide of ammonium to a few drops of a solution of 
 sulphate or nitrate of nickel and boil for a few minutes. 
 Then filter, observe the color of the filtrate, and add dilute 
 hydrochloric acid to acid reaction. 
 
 To illustrate the necessity for the presence of chloride of 
 ammonium to prevent the precipitation of certain members 
 of Class V with the hydrates of Class IV, the following 
 experiments may be performed. Take ten drops of solu- 
 tions of chloride or nitrate of cobalt, dilute with water, then 
 add ammonia- water to alkaline reaction : if any precipitate 
 appear, filter and to the clear filtrate add sulphide of 
 ammonium. Now take the same amount of cobalt solu- 
 tion as before, add three teaspoonfuls of chloride of am- 
 monium solution, then add ammonia-water to alkaline 
 reaction. 
 
CHAPTER VII. 
 
 CLASS VI. — ELEMENTS WHOSE CARBONATES ARE IN- 
 SOLUBLE IN WATER, AMMONIA-WATER AND SALINE 
 SOLUTIONS. 
 
 38. Example of the Precipitation of the Members of 
 
 Class VI. — Place in a small beaker a teaspoonful of aque- 
 ous solutions of the chlorides or nitrates of barium, stron- 
 tium and c alcium . Add to the mixture two or three 
 teaspoonfuls of a solution of ch loride of ammonium , enough 
 a mmonia-water to produce an alkaline reaction, and finally 
 a solution of carbonate of ammoniu m, drop by drop, as long 
 as any precipitate continues to be produced by fresh por- 
 tions of this reagent. To determine this last point, heat 
 the mixture to boiling at intervals, and after boiling allow 
 it to settle until a sufficient quantity of comparatively clear 
 liquid has collected at the top of the mixture to permit the 
 application of the test. 
 
 39. Analysis of the Mixed Carbonates. — The separation 
 of barium, strontium and calcium, one from the other, 
 depends : — 1st. Upon the insolubility of chromate of barium 
 in dilute acetic acid, and the solubility of the chromates 
 of strontium and calcium in that liquid. 2d. Upon the 
 fact that sulphate of strontium is almost absolutely insolu- 
 ble in acidulated water, while sulphate of calcium, though 
 rather sparingly soluble in water, is still sufficiently soluble 
 to be kept in solution. (See App., § 31.) 
 
 Collect the precipitate upon a filter, wash it two or three 
 times with water, taking care to collect the precipitate at 
 the apex of the filter, and dissolve it in acetic acid. The 
 
 65 
 
66 SEPARATION OF BARIUM. [§ 30. 
 
 acid may be poured into the filter as it rests in the funnel, 
 but only a few drops should be used, and the filtrate should 
 be poured back repeatedly upon the filter, until all the pre- 
 cipitate has been dissolved. If the portion of acid at first 
 taken becomes saturated before the precipitate is entirely 
 dissolved, it will be necessary to add an additional small 
 amount. Finally rinse the filter with a little water from 
 a wash-bottle with small orifice, collect the wash-water 
 with the filtrate, and shake the mixture. 
 
 Pour a small portion of the acetic acid solution into a 
 
 test-tube, and add to it a drop of a solution of normal 
 
 chromate of potassium. A pale yellow precipitate falls 
 
 when barium is present, as in this instance; for chromate 
 
 Test of barium is well-nigh insoluble in acetic acid, 
 
 for especially in presence of saline solutions. In 
 
 ^^_: order to separate the whole of the barium, pour 
 
 the contents of the test-tube into the reserved portion of 
 the acetic acid solution, heat the mixture to boiling, and 
 add to it chromate of potassium, until no more precipitate 
 falls and the supernatant liquor appears distinctly yellow, 
 after having been well shaken and allowed to settle. Filter 
 the mixture, and proceed to examine the filtrate for stron- 
 tium and calcium. 
 
 If no barium had been present, no precipitate would have 
 been produced by chromate of potassium in the small por- 
 tion of liquid first tested, and it would have been unneces- 
 sary to mix this reagent with the rest of the acetic acid 
 solution. Trouble would thus be saved, as will appear 
 below. 
 
 It sometimes happens that chromate of barium is precipi- 
 tated in the form of powder so fine that some particles of 
 it pass through the pores of the paper and contaminate the 
 filtrate. Now, in order to detect strontium and calcium it 
 is absolutely necessary that this filtrate, although of a 
 bright yellow color, should be perfectly transparent and 
 
§ 39.] SEPARATION OF STRONTIUM AND CALCIUM. 67 
 
 free from suspended particles of the barium salt. If then 
 the filtrate is at all turbid, it must be poured back re- 
 peatedly into the filter, and again collected in clean tubes, 
 until the last trace of cloudiness has 'disappeared. 
 
 To the filtrate from the chromate of barium add ammo- 
 nia-water to alkaline reaction, and carbonate of ammonium 
 as long as a precipitate falls. Heat the mixture to boiling 
 for a moment, collect the precipitate upon a small filter, 
 and wash it with water, until all the chromate of potas- 
 sium has been removed, and the wash-water runs colorless 
 from the filter. 
 
 Dissolve the precipitate in the smallest possible quantity j 
 of acetic acid, and mix it with three or four times its 
 volume of a solution of sulphate of potassium (App., § 31), 
 made of such strength that, though capable of throwing 
 down sulphate of strontium, it cannot precipitate sulphate 
 of calcium. Allow the mixture to stand at rest for two 
 hours or more, in order that the white powder of sulphate 
 of strontium may separate completely. Then Tests 
 filter, and to the filtrate add ammonia- water to for 
 \y alkaline reaction, and half a teaspoonful of a Sr & Ca. 
 solution of oxalate of ammonium. A white precipitate of 
 oxalate of calcium will be immediately thrown down. 
 
 Since sulphate of strontium is somewhat soluble in a solu- 
 tion of chromate of potassium, the filtrate from chromate 
 of barium cannot be examined directly for strontium by 
 means of sulphate of potassium. The strontium and cal- 
 cium are consequently reprecipitated as carbonates, in order 
 that the excess of chromate of potassium may be washed 
 away. The operation serves also to collect the strontium 
 and calcium out of the mass of liquid in which they have 
 become diffused, and to concentrate them to a small bulk. 
 
 It should be observed that, when the proportion of 
 strontium or calcium in a mixture is small, it often hap- 
 pens that the precipitate, produced by carbonate of ammo- 
 
68 
 
 SEPARATION OF CLASS VL 
 
 [§§ 39-41. 
 
 nium in the filtrate from chromate of barium, is held in 
 suspension and concealed so completely in the yellow- 
 liquor, that an unpractised eye can hardly detect the fact 
 that the liquid ha'fe become cloudy. That a precipitate has 
 really been formed in such cases is easily discovered by 
 throwing a portion of the mixture upon a filter, and com- 
 paring the clear filtrate thus obtained with that portion 
 of the mixture which has been left unfiltered. 
 
 40. An outline of the foregoing operations may be pre- 
 sented in tabulaT form as follows : — 
 
 The General Reagent, [NH4J2C03, of Class VI precipitates the 
 carbonates of Ba, Sr and Ca. Dissolve in dilute acetic acid, and 
 add K^CrO^ : — 
 
 BaCrOi 
 
 is thrown 
 down as 
 a yellow 
 powder. 
 
 Sr and Ca remain in solution. Add (NHJHO and 
 
 (NH4)2C03. Collect and wash the precipitate, and dis- 
 solve it in acetic acid. Add dilute K^SO^ : — 
 
 SrSO, 
 
 is thrown 
 down. 
 
 Ca remains in solution. Add oxalate of 
 ammonium, to precipitate the calcium as 
 
 oxalate. 
 
 i^'t- 
 
 
 41. Separation of Class VI from the Preceding Classes. — 
 
 After Classes I, II, III, IV and V have been separated in 
 the manner already described (§§ 6 to 10), there will still 
 always remain to be examined the filtrate from Class V, 
 and sometimes a precipitate (§ 31) composed of oxalates of 
 barium, strontium, calcium (and magnesium), — in case 
 any salt of these elements, insoluble in ammonia-water, 
 has been thrown down with the members of Class IV. 
 
 If such a precipitate has been obtained in the analysis of 
 Class IV, the oxalic acid contained in it must now be 
 destroyed, the remainder of the precipitate brought into 
 solution, and this solution added to the filtrate from Class 
 
 V, before proceeding to precipitate the members of Class 
 
 VI. To this end, ignite the dry precipitate carefully upon 
 platinum foil, — by several successive portions if the pre- 
 
§ 41.] SEPARATION OF CLASS VL 69 
 
 cipitate is large, — taking care that none of the powder is 
 left sticking to the paper or lost by dropping it from the 
 foil. At a moderate heat the oxalates suifer decomposi- 
 tion, and only carbonates or oxides are left upon the foil. 
 Place the foil and the residue in a small porcelain dish, 
 and dissolve the residue in boiling dilute hydrochloric acid. 
 Add a few drops of chloride of ammonium to the solution, 
 neutralize the acid with ammonia-water, pour the liquid 
 upon a small filter, and add the filtrate to that obtained 
 from Class V. Then add a solution of carbonate of am- 
 monium to the mixture, and boil it in the manner described 
 in § 38. 
 
 If there be no precipitate of the oxalates from Class TV, 
 the filtrate from Class V will, of course, be treated directly 
 with carbonate of ammonium, care being taken to add only 
 a drop or two of the reagent, at first, to ascertain whether 
 any of the members of Class VI are really contained in 
 the solution. 
 
 The solution to which the general reagent c arbonate o f 
 ammonium is added must contain chloride of ammonium, 
 to prevent the precipitation of magnesium a s a carbonat e, 
 and also ammonia-water, to hinder the decomposition of the 
 carbonates of barium, strontium and calcium by the boiling 
 chloride of ammonium. But since the excess of ammonia- , 
 water and the chloride of ammonium, added to the solution 
 before the separation of Class IV, are still contained in it, 
 no new quantity of either of them need here be added. 
 
 One difficulty inherent to this method of analysis is that 
 the carbonates of barium, strontium and calcium are all 
 slightly soluble in a solution of chloride of ammonium. 
 This fact can readily be exhibited by adding a teaspoonful 
 of oxalate of ammonium solution to the filtrate from the 
 precipitate of mixed carbonates (§ 39) and allowing the 
 mixture to stand for some time. A white precipitate of 
 oxalate of barium, strontium or calcium will fall, because 
 
70 SEPARATION OF CLASS VI. [§ 41. 
 
 the oxalates of these metals are less readily soluble than 
 the carbonates in a solution of ammonium chloride. This 
 solubility of the carbonates is so marked that no precipitate 
 whatever is produced, when carbonate of ammonium is 
 added to a weak solution of either of the members of Class 
 VI, in case a large quantity of chloride of ammonium has 
 been previously mixed with it. On this account it is well 
 in an actual analysis to concentrate by evaporation the 
 filtrate from the Class V precipitate before adding the 
 carbonate of ammonium solution, and in some cases, where 
 a very large amount of ammonium salts is present, it is 
 well to evaporate to complete dryness and then to ignite in 
 order to drive off the ammonium compounds. The residue 
 when cold is dissolved in a small quantity of hydrochloric 
 acid ; two or three teaspoonfuls of a solution of chloride of 
 ammonium are added, then ammonia-water, and, finally, 
 carbonate of ammonium, as in § 38. 
 
 In a solution containing traces of barium or strontium 
 these elements might fail to be detected, in case the hydro- 
 chloric acid employed in the process of separating Classes I 
 and II was contaminated with sulphuric acid, or in case the 
 original liquid contained nitric acid to oxidize a portion of 
 the sulphur of the sulphuretted hydrogen employed to pre- 
 cipitate Class II and Class III, or even if the nitric acid, 
 employed to oxidize iron in the filtrate from Classes II and 
 III, were added before the sulphuretted hydrogen had been 
 expelled, or if a powerful oxidizing agent was present in 
 the solution through which sulphuretted hydrogen was 
 passed. Almost all danger is avoided, however, by using 
 pure hydrochloric acid to precipitate Class I, and expelling 
 the nitric acid from the filtrate by evaporating the latter 
 to dryness at a gentle heat, covering the residue with pure 
 concentrated hydrochloric acid, again evaporating to dry- 
 ness, and finally dissolving in water acidulated with hydro- 
 chloric acid. 
 
• CHAPTER VIII. 
 
 CLASS VII. — INCLUDES THE REMAINING COMMON ELE- 
 MENTS NOT COMPRISED IN THE PRECEDING CLASSES, 
 NAMELY: MAGNESIUM. SODIUM AND POTASSIUM. 
 
 42. The Detection of the Several Members of Class VII 
 
 depends: — 1st. Upon the insolubility of a double phos- 
 phate of magnesium and ammonium, and the solubility of 
 the phosphates of x^otassium and of sodium ; and 2d. Upon 
 the fact that compounds of sodium and potassium impart 
 peculiar colorations to non-luminous flames, like those of 
 alcohol and of a mixture of coal-gas and air. 
 
 Prepare a mixture of a small teaspoonful of solutions 
 (App., § 66) of almost any one of the salts of magnesium, 
 sodium and potassium, and add to the mixture an equal 
 bulk of chloride of ammonium. Pour a quarter of the 
 mixture into a test-tube and the remainder into a small 
 porcelain dish. Add to the contents of the test-tube two 
 or three drops of a solution of phosphate of sodium, and 
 as much ammonia-water, and shake the cold mixture at 
 frequent intervals. A crystalline, white precipi- Test 
 tate of the double phosphate of magnesium and for 
 ammonium will appear after a longer or shorter ^S- 
 
 interval, according as the original solution was more or less 
 dilute. 
 
 43. Evaporate the contents of the porcelain dish to dry- 
 ness, ignite the residue until the chloride of ammonium has 
 been completely expelled, — a point which will be indi- 
 cated by the cessation of fuming, — allow the dish to cool, 
 and pour into it three or four drops of water. 
 
 Carefully clean the loop on a piece of platinum wire by 
 
 71 
 
72 SEPARATION OF SODIUM AND POTASSIUM. [§ 43. 
 
 washing it repeatedly with water, and finally holding it in 
 the lamp flame until the last traces of sodium compounds 
 Test ^1*6 burned off, and it ceases to color the flame, 
 for Without touching the loop with*the fingers, dip 
 ^*- it into the aqueous solution in the dish, and again 
 hold it in the flame. A bright yellow color will be im- 
 parted to the flame by the sodium contained in the mixture ; 
 but the color peculiar to potassium compounds will be invis- 
 ible, since the yellow color of the " sodium overpowers and 
 conceals it. Dip the loop a second time in the solution, 
 and again hold it in the lamp flame ; but this time look at 
 the flame through a piece of deep-blue cobalt glass. This 
 cobalt glass is the ordinary blue glass used for stained 
 glass windows; it is essential that the glass should be of 
 moderate thickness, and colored blue throughout, not 
 simply *' flashed" with blue. The characteristic violet 
 Test color imparted to a flame by potassium compounds 
 for will now be visible, for the blue glass shuts off 
 ^' completely the yellow sodium light, while it per- 
 mits the free passage of the violet rays. • 
 
 Since traces of compounds of sodium and potassium are 
 to be found almost everywhere, it is sometimes difficult to 
 determine by the foregoing tests whether the substance 
 under examination contains one or the other of these ele- 
 ments as an essential ingredient, or merely as an accidental 
 impurity. It is always possible, however, to separate the 
 sodium or the potassium from the other members of the 
 class, and to decide, by actual inspection of the isolated 
 compounds, whether one or both of these substances is con- 
 tained in really appreciable quantity in the substance sub- 
 jected to analysis. To this end evaporate the aqueous 
 solution last mentioned which contains the chlorides of 
 magnesium, of sodium and of potassium, almost to dryness, 
 mix thoroughly with the evaporated solution an equal bulk 
 of red oxide of mercury (App., § 54), and ignite the mixture 
 
§§ 43,44.] ISOLATION OF CLASS VIL 73 
 
 until all fuming ceases. The chloride of magnesium will 
 be changed to oxide, while the easily volatile chloride of 
 mercury escapes. 
 
 MgCl2 + HgO = HgCl2 + MgO. 
 
 The ignition should be effected beneath a chimney or in a 
 draught of air powerful enough to carry away the poison- 
 ous fumes of the corrosive sublimate. Boil the residue after 
 ignition with a small quantity of water ; separate the insolu- 
 ble oxide of magnesium together with the excess of oxide 
 of mercury by filtration. E vaporate the filtrate to a small 
 bulk with addition of two or three drops of hy drochlori c 
 acid and a half a teaspoonful of a solution of pl atinic 
 chloride (App., § 56). A yellow crystalline precipitate of 
 chloroplatinate of potassium will separate either at once or 
 after some time. In order to hasten the precipita- xest 
 tion of the potassium compound, which is some- for 
 what less soluble in the presence of alcohol than ^• 
 in water alone, evaporate the solution to which the platinic 
 chloride has been added to a small bulk; disregarding any 
 precipitate "whigh may form, transfer to a test-tube and add 
 to the solution, which should be yellow in color, or more 
 platinic chloride must be added, an equal volume of alcohol ; 
 sha^ and let stand for fifteen minutes. Separate the 
 chloroplatinate of potassium by filtration, and allow the 
 alcoholic filtrate, which should of course still have a yellow 
 color, showing that excess of platinic chloride has Test 
 been added, to evaporate spontaneously in a watch- for 
 glass. Characteristic crystals of chloroplatinate ^^* 
 of sodium will be seen in the form of long, slender prisms 
 or needles of a yellow color. 
 
 44. The Isolation of Class VII, by the removal of the 
 preceding classes, has been described in § 12. Care must 
 always be taken to concentrate the whole of the filtrate 
 from Class VI by evaporation, before testing a portion of 
 
74: SEPARATION OF CLASSES. [§§ 44, 45. 
 
 it for magnesium; and time enough must be allowed for 
 the magnesium precipitate to crystallize ; from very dilute 
 solutions it may not separate for twenty-four hours. A 
 slight flocculent precipitate is often obtained on the addi- 
 tion of the phosphate of sodium to test for magnesium 
 due sometimes to the presence of aluminum, forming phos- 
 phate of aluminum, from the use of too large an excess of 
 ammonium hydrate in Class IV ; sometimes to the imper- 
 fect precipitation of barium, strontium or calcium as car- 
 bonates in the presence of much ammonium chloride or 
 alkaline chlorides. The crystalline character of the mag- 
 nesium compound serves to distinguish it, and can usually 
 be detected without difficulty from the manner in which 
 it forms on the sides of the test-tube, if rubbed with a glass 
 rod, or, if necessary, by using a microscope. 
 
 The remainder of the filtrate from Class VI must be 
 evaporated to dryness and ignited until all fuming ceases, 
 before testing for potassium and sodium, in order that the 
 flame reactions of those elemems may not be concealed or 
 obscured by the vapors of ammonium salts or by the com- 
 bustion of particles of organic matter dei^ived from the 
 various reagents which have been added in the course of the 
 analysis, and also because chloride of ammonium forms with 
 platinic chloride a chloroplatinate of ammonium similar in 
 appearance and insolubility to the potassium compound. 
 In order to apply the chloroplatinate tests for sodium and 
 potassium, it is necessary that they should be present in the 
 solution in the form of chlorides. The iodides of these 
 elements react with platinic chloride to form black platinic 
 iodides, coloring the solution brownish-red in excess of the 
 reagents. 
 
 45. An outline of the methods employed for separating 
 the several classes is here presented in tabular form. The 
 precautions necessary in the consecutive examination of an 
 " unknown " solution for the members of the various classes 
 have been already given under the several classes. 
 
§ 45.] 
 
 SEPARATION OF CLASSES. 
 
 75 
 
 
 e09 
 
 
 
 r^a2 <,o<}^OT o* 9;^^^idioaid Y 
 
 
 
 t 
 
 
 
 p- 
 
 
 ««;> 
 
 <^ 
 
 
 
 tf-^ OOtd^W'^- sa^^^IPai 
 
 
 cc 
 
 p 
 
 
 r^ oj c Oi M. Can 9npis9i y 
 
 loil th 
 ! with 
 
 (See 
 
 p 
 
 
 CD 
 
 P 
 
 % 
 
 
 ^^ ^'B' 
 
 xn 
 
 en 1 
 
 
 Cy^ ^J-CD H 
 
 to S{ 
 
 CD 
 
 
 
 Jr 
 
 
 00 jU^ 
 
 p' 
 
 
 
 *§! 
 
 13 
 
 
 
 ';;i ^ B^ 
 
 ^ 
 
 
 
 WW 
 
 CD 
 
 ^ 
 
 
 s • ^^ U 
 
 ^1 
 
 1 
 
 i 
 
 
 
 ?5 fD 
 
 1 
 
 p 
 
 0' 
 p 
 
 
 ? , IIV 
 
 H 
 
 i>. 
 
 precip- 
 e indi- 
 
 s. 
 
 Co 
 Ni 
 Mil 
 Zn 
 
 e § 35.) 
 
 o 
 
 P 
 
 a- 
 
 CD 
 
 
 
 
 1 
 1 
 
 a- 
 
 en CD 
 
 a* 
 
 
 
 
 H 
 
 •-- 
 
 i«P 
 
 CD 
 
 p- 
 
 
 
 P" 
 
 
 precip- 
 e indi- 
 
 s 
 
 Ba 
 Sr 
 Ca 
 
 e § 39.) 
 
 o 
 
 B 
 
 1 
 
 CD 
 
 
 03 <rt- Hh 130 
 
 
 Evaporate 
 mall bulk, a 
 or Mg. (Se 
 
 Evaporate 
 dryness; 
 md K. (Se 
 
 1 
 
 o 
 
 1 
 
 
 22. 
 
 P-' 
 
 
 
 CO M. e+ CD a r+ 
 
 ?= 
 
 
 
 
 
 
 
 residual 1 
 test a po] 
 42.) 
 
 rest of th 
 ite and tei 
 43.) 
 
 
 1 
 
 CO 
 
 
 i 
 
 
 
 ^ ^ dS 
 
 p 
 
 
 H 
 
 Ui 
 
 
 
 ^Oi op 
 
 w 
 
 
 
 
 
 
 o o P o 
 
 
 
 CD 
 
 CD 
 
 
 
 ^ P r^ "" 
 
 
 
 P 
 
 COS 
 
 
 
 SjI S,^ 
 
 
 
 t 
 
 .^ 
 
 
 
 » p ?;-p 
 
 
 
 P^ 
 
 c^ 
 
 
CHAPTER IX. 
 
 IRAL TESTS FOR THE NON-METALLIC ELEMENTS. 
 
 46. The following common elements remain to be con- 
 sidered : — Sulphur, nitrogen, phosphorus, carbon, boron, 
 silicon, chlorine, bromine, iodine, fluorine, oxygen, hydro- 
 gen. It is obvious that oxygen and hydrogen cannot be 
 directly sought for by any analytical process conducted in 
 the wet way. These elements are added to the original 
 matter in the water or acid which is used as a solvent. 
 The presence of these elements is either inferred from the 
 other results of the analysis, or forms the object of direct 
 inquiry in the preliminary treatment, — that important part 
 of every analysis which forms the main subject of Part II. 
 The remaining elements all belong to that class vaguely 
 described as non-metallic; they unite with oxygen, hydro- 
 gen, or both these elements, to form what are commonly 
 called adds. 
 
 As the metallic elements are recognized through familiar 
 compounds, hydrates, chlorides, sulphides, or other, so 
 these non-metallic elements are identified by working out 
 of the unknown mixture their characteristic compounds. 
 These compounds are generally salts. But there is one 
 marked difference between the search for the metallic and 
 the search for the non-metallic elements in the wet way. 
 In the case of iron and mercury it is commonly necessary 
 to determine whether the element, if present, exists as a 
 ferrous or ferric, as a mercurous or mercuric salt, in the 
 case of chromium whether as a chromate or a salt of chro- 
 mium; but as a rule it is true that no question usually 
 76 
 
§ 46.] CLASSES OF SALTS, 77 
 
 arises in connection with the determination of a metallic 
 element, except the fundamental one of its presence or 
 absence in a given solution. The sodium in the three 
 different salts, sulphate, sulphite and hyposulphite of 
 sodium, for example, is detected by one and the same 
 method; but, when the other elements of these salts 9jjj^l 
 sought for, three different reactions will occur, accordin^^^' 
 the varying nature of the ingredients other than sodium. 
 A sulphate in solution gives one set of reactions, a sulphite 
 another, and a hyposulphite a third. It is important to 
 do more than determine the mere presence of sulphur and 
 oxygen. We want to know whether the sodium salt be a 
 sulphate, sulphite or hyposulphite. Analogous questions 
 arise with regard to other non-metallic elements; there is 
 chlorine both in chlorides and chlorates, and carbon both in 
 carbonates and oxalates. Arsenic, too, may be present in 
 the form of arsenite or arseniate, and these two different 
 kinds of salts exhibit many quite dissimilar reactions. 
 
 In the systematic analysis of an unknown substance, the 
 examination for the metallic elements by the methods 
 detailed in the previous chapter, as a ruUy precedes the 
 determination of the non-metallic elements. Having ascer- 
 tained which of the metallic elements are present in the 
 substance under examination, the analyst tries to identify 
 each class of salts which may be present by precipitating, 
 or otherwise making manifest, some familiar member of 
 that class. Thus he identifies sulphates by precipitating 
 sulphate of barium, chlorides by precipitating chloride of 
 silver, carbonates by throwing down carbonate of calcium, 
 and so forth. The practically important inquiry is, how to 
 find means of identifying each of the principal classes or 
 kinds of salts. The word "class" or "kind" of salts is 
 used in this connection as a collective name for all the 
 salts which may be regarded as derived from one and the 
 same a,cid; thus all sulphates constitute one class or kind, 
 

 78 TESTING FOE NON-METALLIC ELEMENTS. [§ 46. 
 
 all carbonates another, and so forth. The following com- 
 mon classes of salts are those for which more or less per- 
 fect means of recognition will be given in this chapter : — 
 Sulphates, sulphites, hyposulphites, sulphides, arseniates, 
 arsenites, phosphates, carbonates, oxalates, tartrates, bo- 
 
 es, silicates, chromates, fluorides, chlorides, bromides, 
 des, cyanides, nitrates, chlorates, acetates. 
 
 No system of successive testing and elimination, analo- 
 gous to that already described for the metallic elements, has 
 been devised for the non-metallic constituents of salts. 
 There are, indeed, certain so-called general reagents for 
 adds; but these reagents are chiefly useful to show the 
 simultaneous presence or absence of members of several 
 classes of salts, and hardly help towards the identification of 
 any individual class, except in cases of the simplest sort, 
 in which only a single class of salts is represented. 
 
 It sometimes happens that the preliminary examination 
 (§§ 82, 83) of the substance to be analyzed leads directly to 
 some conclusion as to the class of salts in hand. A knowl- 
 edge of the metallic element or elements present is more- 
 over in most cases a great help in the determination of the 
 other constituents. It is for this reason that the search 
 for the metallic elements precedes the examination for the 
 non-metallic. A single example will sufficiently illustrate, 
 for the present, this important principle, which is of very 
 wide application in qualitative analysis. 
 
 Suppose that the substance under examination is a solid 
 which dissolves without difficulty in water and which 
 proves to contain barium. From the presence of barium 
 in a compound soluble in water, it is to be directly inferred 
 that there is no sulphate, phosphate, carbonate, oxalate or 
 tartrate in the original substance since these barium salts are 
 insoluble in water. The list of salts excluded by the demon- 
 strated presence of barium would be very long. So it is in 
 greater or less degree with many other metallic elements. 
 
|§ 46, 47.] THE BARIUM TEST. 79 
 
 It is, indeed, no easy matter to make a solution containing 
 even one representative of the seven classes into whicli the 
 metallic elements have been divided, because each of these 
 elements, except sodium and potassium, when present in 
 a solution, excludes one or more whole classes of salts. By 
 attending to just inferences to be drawn from the quality 
 of the metallic elements, much time will be saved, and the 
 want of a systematic procedure in searching for the non- 
 metallic elements will be little felt. The presence of the 
 arsenites and arseniates, of carbonates, chromates, cyanides, 
 sulphides, sulphites and hyposulphites will ordinarily be 
 revealed in the course of the search for the metallic 
 elements. 
 
 Before giving the special tests by which the above-men- 
 tioned classes of salts are identified, we proceed to describe 
 certain general tests which are of value, particularly when 
 they give negative results. 
 
 47. The Barium Test. — When a solution of chloride (or 
 nitrate) of barium is added in suitable quantity to a neutral 
 or slightly alkaline solution containing representatives of 
 any or all of the following classes of salts, precipitation 
 "usually occurs, for all these salts of barium are insoluble in 
 water and alkaline liquids, if no ammonium salts be pres- 
 ent : — 
 
 Sulphates, Phosphates, Silicates, 
 
 Sulphites, Carbonates, Chromates, 
 
 Hyposulphites, Oxalates, Fluorides. 
 
 Arseniates, Tartrates, 
 
 Arsenites, Borates, 
 
 If the chloride (or nitrate) of barium fail to produce a 
 precipitate under the prescribed conditions, the complete 
 absence of all the above thirteen classes of salts is at once 
 demonstrated, provided that no ammonium salts be con- 
 tained in the original solution. 
 
80 THE BARIUM TEST, [§§ 48, 49. 
 
 48. Illustration of the Barium Test. — Prepare in a test- 
 tube a solution containing at once sulphate, phosphate and 
 carbonate of sodium ; a very small bit of each salt will be 
 sufficient. The solution will be found to be alkaline to 
 litmus paper. Add to it chloride of barium (App., § 44), 
 little by little, until a fresh addition of the reagent no 
 longer produces an additional precipitate. The white pre- 
 cipitate consists of sulphate, phosphate and carbonate of 
 barium. All the thirteen barium salts which are liable to 
 precipitation under these circumstances are white, with the 
 single exception of chromate of barium. The yellow color 
 of the chromate of barium (§ 39) distinguishes this one 
 precipitate from all the rest. If this color is well marked, 
 the presence of a chromate in the original solution (which 
 must also have been yellow) may be inferred with cer- 
 tainty. In all other cases, however, the precipitate is 
 white, as in the present experiment. The next question 
 is, can anything further be learned from this white precipi- 
 tate? 
 
 Add to the contents of the test-tube dilute hydrochloric 
 acid, until the liquid has a decidedly acid reaction to lit- 
 mus paper. An effervescence indicates the escape of car- 
 bonic acid, displaced by the less volatile hydrochloric acid. 
 The bulky precipitate which the chloride of barium pro- 
 duced will in part disappear, but a portion of it remains 
 undissolved. Filter the contents of the tube. The parti- 
 cles of precipitated sulphate of barium are so very fine that 
 they often pass through the pores of the paper, necessitat- 
 ing repeated filtration through the same filter. To the 
 filtrate add ammonia-water until the liquid has an alkaline 
 reaction. A precipitate will reappear: the phosphate of 
 barium which was dissolved by the hydrochloric acid is 
 reprecipitated as soon as the acid solvent is neutralized by 
 ammonia- water. 
 
 49, Of all the barium salts which might, in an actual 
 
§ 49.] THE BARIUM TEST. 81 
 
 analysis, be precipitated under conditions similar to those 
 of the preceding experiment, only one, the sulphate of 
 barium, is insoluble in dilute hydrochloric acid. A sepa- 
 ration of sulphur from a hyposulphite will not be mistaken 
 for a barium precipitate (§ 19, p. 22). The fact that any 
 of the original precipitate remains undissolved by the 
 hydrochloric acid demonstrates the presence of a sulphate. 
 The portion of the original precipitate which dissolved in 
 the acid, and was reprecipitated by ammonia, consisted in 
 this particular experiment of phosphate of barium; but 
 in an actual analysis the possible salts represented would 
 be so numerous as to make the indication of but little value ; 
 on the other hand, if ammonia-water causes no precipita- 
 tion, it must not be inferred that the acid of course dis- 
 solved nothing, for, with the exception of the sulphate, all 
 the thirteen above-mentioned barium salts are more or less 
 soluble in solutions of ammonium salts, and if present in 
 small quantity may not be precipitated on addition of am- 
 monia. This is especially true of the borate, oxalate, 
 arseniate, arsenite, tartrate and fluoride, and when ammo- 
 nium -salts are known to be present, there is really but one 
 perfectly satisfactory indication with chloride of barium. 
 TJie presence of a sulphate is revealed by it with certainty, but 
 the results of the other tests, if negative, must be received 
 with some distrust. 
 
 It is to be said, however, that if the substance under 
 examination is known to be a simple salt and is found to 
 be a salt of ammonium, the barium test can hardly fail to 
 give a precipitate in alkaline solution, if the unknown 
 compound is actually a borate, or phosphate, etc., of ammo- 
 nium. 
 
 If the original solution be acid, it is necessary to neutral- 
 ize it with ammonia-water before the chloride of barium is 
 added. If a precipitate is produced, it is necessary to filter 
 and proceed with the filtrate. The precipitate may contain 
 
82 THE CALCIUM TEST. [§§ 49, 50. 
 
 phosphates, oxalates, etc., as explained in § 29, page 46, 
 and this must be ascertained by special tests, §§ 66, 67, 69, 
 etc. Even if ammonia-water produce no precipitate, the 
 testing is still performed under the disadvantage of the 
 presence of ammonium salts. 
 
 If the original solution contained silver or lead salts, or 
 mercurous salts, it would be impossible to use chloride of 
 barium and hydrochloric acid as reagents; they would 
 throw down the chlorides of those metals. Nitrate of 
 barium (App., § 45) and dilute nitric acid must then be 
 used. 
 
 The acids added to the precipitate formed by the barium 
 salt must be always dilute acids. Chloride and nitrate of 
 barium are themselves insoluble in concentrated hydro- 
 chloric and nitric acids, and if a strong acid were used as 
 a solvent, the reagent salt might itself separate from the 
 liquid. 
 
 50. The Calcium Test. — Chloride (or nitrate) of calcium 
 precipitates the same classes of salts as chloride (or nitrate) 
 of barium, with* the single exception of the chromates. 
 When sulphates and all ammoniacal salts are absent, or pres- 
 ent only in minute quantities, something may be learned 
 by testing a neutral or slightly alkaline solution supposed 
 to contain representatives of some of the other classes of 
 salts enumerated in § 47, with chloride or nitrate of calcium. 
 The calcium salts liable to precipitation under these cir- 
 cumstances are all soluble in acetic acid, except the oxalate 
 and the fluoride. The precipitate jjroduced by the calcium 
 reagent is, therefore, treated with acetic acid; if it com- 
 pletely redissolves, oxalates and fluorides are most probably 
 absent. The presence of notable quantities of ammonium 
 salts renders this test of uncertain value, because the 
 fluoride and many other salts of calcium are soluble in 
 solutions of ammonium salts. Since sulphate of calcium 
 is sparingly soluble in water and acetic acid, the presence 
 
§§ 60-53.] TH:E StLVEU TEST. . 88 
 
 of a sulphate, causing precipitation of sulphate of calcium, 
 obscures the reaction for oxalates and fluorides. Nitrate of 
 calcium must be used instead of the chloride whenever 
 silver or lead salts, or mercurous salts, are present in the 
 solution under examination. 
 
 51. Illustration of the Calcium Test. — Prepare in a test- 
 tube an aqueous solution of phosphate, oxalate and tartrate 
 of sodium or potassium. A very small quantity of each 
 salt will be enough. The solution will be neutral or faintly 
 alkaline. Add to this solution a solution of chloride of 
 calcium (App., § 43) until the precipitation is complete.' 
 Collect the wliite precipitate upon a filter, and, when 
 drained, transfer it to a test-tube and treat it with acetic acid. 
 The phosphate and tartrate of calcium will redissolve, 
 but the oxalate remains untouched. To verify this result, 
 and identify each one of the classes of salts present in the 
 original solution, special tests, to be hereafter described, 
 must be resorted to. 
 
 62. The Silver Test. — Nitrate of silver produces a pre- 
 cipitate in neutral or acid solutions with all chlorides, 
 bromides, iodides and cyanides, and in neutral solutions 
 with most of the classes of salts enumerated in § 47. In 
 
 . order to obtain the most comprehensive negative conclusion 
 in the case that nitrate of silver produces no precipitate, it 
 is necessary to operate upon a rieiitral solution. If, on the 
 addition of nitrate of silver to a neutral solution, no precipi- 
 tate appears after the lapse of several minutes, neither 
 chlorides, bromides, iodides, cyanides nor sulphides can be 
 present, and the absence of sulphites, hyposulphites, car- 
 bonates, phosphates, arseniates, arsenites, chromates, sili- 
 cates, oxalates and tartrates may also be inferred with 
 considerable certainty. 
 
 63. Illustration of the Silver Test. — Prepare in a test- 
 tube a weak solution of chloride of sodium, iodide of 
 
 'potassium, cyanide of potassium and phosphate of sodium. 
 
84 THE SILVER TEST. [§§53,64. 
 
 Add nitrate of silver (App., § 40) to this slightly alkaline 
 solution, until the precipitation is complete. The dense 
 precipitate is yellowish-white. Pour dilute nitric acid 
 into the mixture, until the solution is strongly acid; shake 
 up the contents of the tube thoroughly, and after the lapse 
 of several minutes collect the insoluble precipitate on a 
 filter, and receive the filtered liquid in a test-tube. 
 
 Neutralize the filtrate with ammon ia- water : a yellow 
 precipitate of phosphate of silver will reappear. 
 
 Wash the precipitate on the filter thoroughly to free it 
 from the superfluous nitrate of silver. Einse the washed 
 precipitate into a clean test-tube, decant the water from 
 above it, pour over it ammonia-water, and gently heat the 
 mixture. The silver precipitate will visibly diminish in 
 bulk, but a yellowish portion remains undissolved. Filter 
 again, and neutralize the filtrate with nitric acid; a white 
 precipitate will fall. 
 
 This experiment proves that a portion, but not all, of 
 the mixed silver salts which are insoluble in nitric acid, 
 are soluble in ammonia-water. The chloride, cyanide and 
 bromide of silver dissolve in ammonia-water, the latter with 
 difficulty; the iodide remains undissolved. Special tests, 
 hereafter to be described, are applied in order to confirm the 
 presence of iodine, and to detect each and all of the three 
 substances which are liable to be confounded in the white 
 precipitate just mentioned. 
 
 54. In the application of the silver test to the examina- 
 tion of a substance of unknown composition, it is, of course, 
 most advantageous to work with a neutral solution, as in 
 such a case the absence of any precipitate would lead to the 
 inference that all the salts mentioned in § 52 are absent : if 
 the solution is neutral, the nitrate of silver may be added 
 directly to a portion of it. In addition to the classes of 
 salts mentioned in § 52, if the original solution contained 
 any considerable quantity of a borate, the borate of silver 
 
§ 51.] THE SILVER TEST. • 85 
 
 would be precipitated under these conditions ; but a small 
 portion of some borate might escape precipitation. If the 
 original solution be acid to test paper, add nitrate of silver 
 to a portion of it in a test-tube, and then pour in upon the 
 liquid some dilute ammonia-water, so gently that the two 
 liquids do not mix at once. At some layer near the junc- 
 tion of the two dissimilar liquids, the fluid must be neutral. 
 If at that layer no precipitate or cloud appear, the twelve 
 kinds of salts above enumerated are absent. If the original 
 solution is alkaline, dilute nitric acid is to be added in 
 precisely the same manner as the ammonia-water in the 
 opposite case. The neutral layer between the two liquids 
 is attentively observed, and the absence of any film or 
 cloud therein justifies the same sweeping conclusion as that 
 above given. 
 
 Some conclusions may often be drawn from the color of 
 the precipitate produced by nitrate of silver. Chloride, 
 bromide, cyanide and oxalate of silver are white; tartrate 
 white becoming black on boiling; borate white, though 
 normal borates form in part brown silver oxide; iodide, 
 phosphate and arsenite of silver are yellow ; silicate of sil- 
 ver is yellow or white; arseniate of silver is brownish-red; 
 chromate of silver is purplish-red; sulphide of silver is 
 black. When the silver precipitate is white, black, or of 
 some obscure, indecisive color, the operations in the wet 
 way at this stage should be directed to proving or disprov- 
 ing the presence of chlorides, bromides, iodides, cyanides 
 and sulphides. To this end the portion of the original 
 solution which has been already tested with nitrate of sil- 
 ver should be made decidedly acid with dilute nitric acid. 
 Of all the silver salts which can be precipitated on the 
 addition of nitrate of silver, only the chloride, bromide, 
 iodide, cyanide and sulphide can resist dilute nitric acid. 
 If the precipitate once formed redissolves completely in 
 nitric acid, no chloride, bromide, iodide, cyanide or sul- 
 
86 * THE SILVER TEST. 
 
 pliide was present in the original solution. If, on the con- 
 trary, a residue remain, one or more of these kinds of salts 
 must have been represented in the original solution. If 
 the residue be black or blackish, the presence of a sulphide 
 is to be inferred; if it be white or whitish, the absence of 
 sulphides and the presence of a chloride, bromide or cyanide 
 is to be inferred; if it be distinctly yellowish, an iodide is 
 probably present. When the liquid under examination 
 contains a ferrous salt, protochloride of tin, or any other 
 active reducing agent, metallic silver is liable to be precip- 
 itated as a dark, heavy powder. The examination for the 
 metallic element will generally have revealed the presence 
 of any such agent. 
 
 It is to be remarked that cyanide of mercury does not 
 give a precipitate with nitrate of silver. When mercury 
 has been detected in the substance under examination, 
 cyanogen must be sought for in other ways (§ 59) than this. 
 
 55. Nitrates, Chlorates and Acetates. — It is quite clear 
 that no method of precipitation whatever will apply to 
 nitrates, chlorates and acetates, since no insoluble chlorate 
 is known, and, with the exception of some rather ill-defined 
 basic compounds, the nitrates and acetates are also all 
 soluble. Special tests must be sought for these three classes 
 of salts. 
 
 It is obvious that it is only in the case of the analysis of 
 substances soluble in water with neutral or slightly alkaline 
 reaction that the general tests are of the greatest value. 
 
CHAPTER X. 
 
 SPECIAL TESTS FOR THE NON-METALLIC ELEMENTS. 
 
 56. We now proceed to indicate the most available special 
 tests for the non-metallic elements and their commonest 
 compounds. It is noteworthy that the non-metallic ele- 
 ments enter into composition under various forms, which 
 produce, with one and the same metallic element, various 
 salts. Thus within the narrow range of this treatise, sul- 
 phur is to be sought in sulphides, sulphates, sulphites and 
 hyposulphites (thiosulphates) ; carbon in cyanides, acetates, 
 carbonates, oxalates and tartrates; and arsenic in arsenites 
 and arseniates. The various classes of salts will be taken 
 up successively. It should be premised that these special 
 tests are sometimes applied to the original solution before 
 the precipitation with chloride of barium and nitrate of 
 silver, and sometimes during or after these more general 
 testings. In the first case, the student is seeking guidance 
 in the application of the more comprehensive tests; in the 
 latter case, he is trying to confirm results already almost 
 sure. 
 
 57. Effervescence. — When a solution containing a carbo- 
 nate, cyanide, sulphide, sulphite, or hyposulphite, or a mix- 
 ture of representatives of some or all of these kinds of 
 salts, is treated with hydrochloric acid and then warmed, an 
 evolution of gas occurs with more or less effervescence. The 
 gases evolved are all colorless; but they all have very 
 characteristic odors except carbon dioxide, the gas which 
 escapes from a carbonate. A cyanide gives off the pungent 
 odor of hydrocyanic acid. A sulphide yields sulphuretted 
 
 87 
 
8d CARBONATES. — CYANIDES. [§§57-59. 
 
 hydrogen of familiar presence. Sulphurous acid gas (SO^) 
 escapes from sulphites and hyposulphites (thiosulphates) 
 alike; but in the latter case a deposition of sulphur makes 
 the liquor turbid. If only one of these gases were present, 
 the effervescence and the odor, or the absence of odor, 
 would identify it; but when mixtures are to be dealt with, 
 further means of identification are necessary. 
 
 68. Carbonates. — To prove the presence of carbonic acid 
 gas, when effervescence has occurred, add h yrlrnp.hloric acid , 
 little by little, to the effervescing solution, until the acid is 
 decidedly in excess ; meanwhile keep the mouth of the test- 
 tube loosely closed with the thumb to promote the accumu- 
 lation of the gas evolved. When the tube is supposed to 
 be full, carefully decant the gas into a second test-tube 
 containing a teaspoonful of li me-water (App., § 42), taking 
 care not to allow any of the liquid to pass over with the 
 gas. Mix the lime-water and the gas in the second test- 
 tube by thorough shaking. A white precipitate of carbo- 
 nate of calcium will be produced, if the gas tested is, or 
 contains, carbonic acid gas (COj). 
 
 If the effervescence is slight, and the quantity of gas 
 evolved seems too small to be decanted in this way, dip the 
 end of a dark-colored glass rod into lime-water, and thrust 
 the moistened end into the test-tube, bringing it close to the 
 surface of the fluid. If the gas be carbonic acid gas the 
 lime-water adhering to the rod will become visibly turbid. 
 
 The student who desires to see the working of this test 
 before having occasion to apply it in an actual analysis, 
 can operate upon a morsel of carbonate of sodium dissolved 
 in a little water. 
 
 59. Cyanides. — When the smell of the gas which escapes 
 from the solution under examination (§ 57), or the qualities 
 of the precipitate wjth nitrate of silver (§ 54) give occasion 
 to suspect the presence of a cyanide, the following confirm- 
 atory test may be resorted to : — add to the solution sup- 
 
§§ 59, 60.] SULPHIDES. 89 
 
 posed to contain free hydrocyanic acid or an alkaline 
 cyanide, a few drops of a solution containing both a ferrous 
 and a ferric salt (a solution of ferrous sulphate which has 
 been exposed to the air, for example) and a small quantity 
 of sodium hydrate solution. A precipitate forms, which, if 
 cyanogen is present, consists of a mixture of Prussian blue 
 and the hydrates of iron. Warm the liquid, and add to it 
 excess of hydrochloric acid ; the hydrates of iron dissolve, 
 but the Prussian blue remains undissolved. If the amount 
 of cyanogen is small, the hydrates of iron, which of course 
 always form, may conceal the Prussian blue until the addi- 
 tion of the acid; and in the case of minute quantities the 
 liquid simply appears gj-een after the addition of hydro- 
 chloric acid, and it is only after long standing that a tri- 
 fling blue precipitate separates from it. 
 
 This test may be well illustrated by means of a small 
 particle of cyanide of potassium dissolved in a teaspoonful 
 of water. 
 
 To detect cyanogen in cyanide of mercury, it is necessary 
 to precipitate the mercury as sulphide, by means of sulphu- 
 retted hydrogen, and to identify the free hydrocyanic acid 
 in the filtered or decanted liquid. 
 
 60. Sulphides. — Many sulphides give off sulphuretted 
 hydrogen when heated with hydrochloric acid. If the 
 quantity of gas is so small that its odor is imperceptible, 
 the lead paper test (§ 33) should be applied. 
 
 When sulphides are dissolved in nitric acid or aqua regia, 
 their sulphur is partly separated in a free state, and partly 
 converted into sulphuric acid. The free sulphur can be 
 identified by its color and texture, and by its behavior 
 when burnt. The sulphuric acid in the liquid is detected 
 in the usual manner (§ 65). 
 
 If on warming the substance to be tested with hydrochlo- 
 ric acid other than a white residue remains, it may contain 
 sulphides, although no test for sulphuretted hydrogen is 
 
90 SULPBITES. — HYPOSULPHITES. [§§ 60^2. 
 
 obtained: in this case, warm gently for a few minutes, add 
 a granule of zinc, warm, and repeat the test with lead paper. 
 If no evidence of sulphuretted hydrogen is obtained, and 
 a colored residue still remains, remove the zinc, wash the 
 residue several times by decantation to remove soluble 
 sulphates, treat with aqua regia, warming until no further 
 action takes place, dilute with eight or ten parts of water, 
 filter, and to the clear filtrate add a little chloride of 
 barium solution (nitrate of barium, if lead or silver have 
 been found in the search for metallic elements); a white 
 precipitate of sulphate of barium shows that a sulphide or 
 free sulphur was originally present. 
 
 61. Sulphites. — All the sulphites evolve sulphurous acid 
 gas (SO^), without any deposition of sulphur, when treated 
 with hydrochloric acid. The very sharp odor of this gas 
 is enough to identify it. Traces of sulphurous acid gas 
 may be readily detected by exposing to the gas in the open 
 end of a small tube a drop of a mixture of ferric chloride 
 solution and ferrocyanide of potassium, Prussian blue being 
 formed — resulting from the reduction of the ferric com- 
 pound. 
 
 As has been before stated, nitrate of silver produces in 
 solutions of sulphites a white precipitate ; but this precipi- 
 tate blackens when the liquid is boiled, on account of the 
 reduction of silver. 
 
 When the sulphites are heated with strong nitric acid, or 
 other powerful oxidizing agent, they are converted into 
 sulphates without precipitation of sulphur. The sulphates 
 so produced may be identified in the usual way (§ 65). 
 
 62. Hyposulphites (thiosulphates) . — The hyposulphites 
 disengage sulphurous acid gas (SO2) and deposit sulphur 
 when warmed with hydrochloric acid. This decomposition 
 is not immediate if the solution be dilute. The precipitated 
 sulphur is yellow, and not white, as is usually the case when 
 sulphur separates in chemical reactions. 
 
§§ 62-64.] CHROMATES. — ARSENITES, AESENIATES. 91 
 
 The precipitate produced in the solution of a hyposulphite 
 by nitrate of silver dissolves again readily in an excess of 
 the hyposulphite. On standing, the precipitate of hyposul- 
 phite of silver turns black spontaneously,*being decomposed 
 into sulphide of silver and sulphuric acid. Heating pro- 
 duces this eifect almost immediately. 
 
 Hyposulphite of sodium is a good salt from which to get 
 the above reactions of hyposulphites. 
 
 63. Chromates. — All chromates are yellow iii color; 
 bichromates are orange-red. Chromium is detected during 
 the examination for the metallic elements, and the analyst 
 generally obtains pretty certain evidence concerning the 
 actual condition in which this element enters into the sub- 
 stance under examination, whether as chromate or salt of 
 chromium ; for the chromates are reduced by sulphuretted 
 hydrogen with change of color and deposition of sulphur, 
 as stated in § 23 c. Unless reduced by sulphuretted hydro- 
 gen, no precipitate of chromium as hydrate will be formed 
 in Class IV. 
 
 To confirm the indications thus obtained, the following 
 tests are used: — When acetate of lead (App., § 46) is 
 added to a neutral solution of a chromate, yellow chromate 
 of lead separates, insoluble in acetic acid, but soluble in 
 caustic soda. 
 
 As has been before stated, the purplish-red color of 
 chromate of silver (§ 54), and the yellow color of chromate 
 of barium (§ 49), are valuable indications of the presence 
 of a chromate. 
 
 A solution of normal chromate of potassium is the best 
 substance from which to obtain for the first time the reac- 
 tions of the chromates. 
 
 64. Arsenites and Arseniates. — The presence or absence 
 of arsenic is determined in the search for the metallic 
 elements by the tests already given in the treatment of 
 Class HI. These tests, however, do not indicate which 
 
92 ARSENITES AND ARSENIATES. [§ 64. 
 
 class of compounds (arsenic or arsenious) is present in the 
 solution under examination. In whatever form present, 
 the arsenic would be precipitated as the trisulphide (AS2S3), 
 and subsequently^converted into an arseniate by fusion with 
 nitrate of sodium. Discriminating tests must therefore be 
 applied to the original solution. 
 
 The tests by which an arseniate may be identified have 
 already been given (§ 25, p. 35). It may be remarked 
 further that the presence of this class of arsenic compounds 
 is often inferred by the length of time required to saturate 
 with sulphuretted hydrogen a solution containing arsenic 
 acid or an arseniate. The arsenic acid is first reduced to 
 arsenious acid and then precipitated; a separation of sul- 
 phur accompanies the reduction. Unless the solution is 
 hot the arsenic may not be precipitated as sulphide^ for 
 several hours. 
 
 A solution of an arsenite yields with nitrate of silver a 
 yellow precipitate of arsenite of silver under the same con- 
 ditions in which an arseniate gives the brownish-red precipi- 
 tate. The silver test generally serves to distinguish between 
 arsenites and arseniates, but circumstances may arise in 
 which it would be inapplicable. The following test affords 
 further means of discrimination : — 
 
 If to a solution of arsenious acid or an arsenite, caustic 
 soda be first added in excess, and then five or six drops of 
 a dilute solution of sulphate of copper, a clear bluish liquid 
 is obtained, which, upon boiling, deposits a red precipitate 
 of suboxide of copper (CU2O), while soluble arseniate of 
 sodium is simultaneously produced, and remains in the 
 solution. This test is good as a means of distinguishing 
 between arsenites and arseniates when no organic matters 
 are contained in the solution under examination. The 
 qualification is necessary, because grape-sugar and many 
 other organic substances exercise a like reducing action on 
 copper salts. 
 
§§64-66.] SULPHATES. —PHOSPHATES. 93 
 
 In case copper salts are present in the solution to which 
 the silver test for arseniates and arsenites is applied, there 
 is oftentimes difficulty in applying the test. When copper 
 has been found in the search for metallic elements, the fol- 
 lowing method may be employed. Boil a portion of the 
 original substance for fifteen or twenty minutes with three 
 or four teaspoonfuls of a saturated solution of sodium car- 
 bonate, adding water from time to time as it evaporates; 
 filter, and acidify the clear filtrate with nitric acid, add 
 nitrate of silver solution, and then neutralize with ammonia- 
 water : the precipitate of arseniate, or arsenite, of silver will 
 appear. If the ammonia-water is added carefully so as not 
 to mix with the acid solution, the precipitate will appear 
 as a thin film of characteristic color near the junction of 
 the two dissimilar liquids. 
 
 65. Sulphates. — The barium test (§§ 47-49) is all-suffi- 
 cient for the detection of sulphates. Sulphate of barium 
 being the only s alt of barium insoluble in dilute acids , when 
 a white precipitate appears on the addition of a solution 
 of chloride of barium to a solution acidified with hydro- 
 chloric or nitric acid, the presence of sulphuric acid or a 
 sulphate is shown. In case lead, silver or mercury in the 
 mercurous condition are present, nitrate of barium solution 
 must be used instead of chloride. It should be remembered 
 that both chloride and nitrate of barium are themselves 
 somewhat insoluble in strongly acid solutions; also that 
 common nitric and hydrochloric acids always contain traces 
 of sulphuric acid. 
 
 66. Phosphates. — When two or three drops of a neutral 
 or n itric acid so lutio n of a phosphat e (even of iron, alumi- 
 num, barium, strontium, calcium or magnesium [compare 
 § 29], are poured into a test-tube containing four or five tea- 
 spoonfuls of a solution of molybdate of^ammonium in nitric 
 acid (App., § 22), there is formed a pale yellow precip- 
 itate which is apt to gather upon the sides and bottom 
 
94 PHOSPHATES. [§ 66. 
 
 of the tube. The test is best applied to a nitric acid solu- 
 tion of the substance ; shaking vigorously after addition to 
 the molybdate of ammonium solution, and if necessary, 
 warming gently (not above 70° C). If the precipitate does 
 not appear in a few minutes, a few drops more of the solu- 
 tion to be tested may be added. Let the mixture stand for 
 at least half an hour to detect traces of phosphoric acid. 
 
 This precipitate is soluble in an excess of phosphoric and 
 other acids; and certain organic substances may prevent 
 its formation. A yellow coloration of the liquid merely is 
 not enough to prove beyond question the presence of a 
 phosphate ; a precipitate must be waited for. The yellow 
 precipitate can be easily recognized, even in dark-colored 
 liquids, when it has settled. 
 
 When the previous steps of the analysis have proved that 
 the phosphates present in the solution under examination 
 are soluble in ammoniacal liquids, and that no arsenic acid 
 or arseniates are present, the following test will identify a 
 phosphate or free phosphoric acid : — 
 
 - Add to the solution to be tested a clear mixture of chloride 
 of magnesium, chloride of_ ammonium and ammonia- water 
 (App., fif). When a phosphate or free phosphoric acid 
 is present, a white crystalline precipitate of phosphate of 
 magnesium and ammonium is formed, even in very dilute 
 solutions. Stirring and shaking promote its separation. 
 The precipitate dissolves readily in acids. 
 
 When arsenic acid or arseniates are present in the origi- 
 nal mixture, this test for phosphates can still be applied, 
 if all the arsenic be previously removed by precipitation as 
 sulphide (§ 25). The magnesium mixture can be used in 
 the filtrate from the sulphide of arsenic, after it has been 
 boiled to expel the sulphuretted hydrogen. 
 
 The magnesium test can only be applied when the phos- 
 phate present is soluble in ammoniacal solutions. 
 
 Phosphate of sodium is the best substance on which to 
 try the test for phosphates. 
 
§§ 67, 68.] OXALATES. — TARTRATES, 95 
 
 67. Oxalates. — The precipitation of white, finely divided 
 oxalate of calcium, by all soluble calcium salts from solu- 
 tions of oxalates or oxalic acid, has been already described 
 (§ 51). Even the solution of sulphate of calcium gives this 
 reaction with oxalates. 
 
 If oxalic acid, or an oxalate in the dry state, be heated 
 in a test-tube with an excess of concentrated sulphuric acid, 
 a mixture of carbonic oxide and carbonic acid is set free 
 with effervescence ; the carbonic acid may be identified by 
 the lime-water test. Provided that no carbonate is pres- 
 ent in the substance to which the test is applied, and if the 
 quantity operated upon is considerable, the carbonic oxide 
 may be inflamed at the mouth of the tube. 
 
 68. Tartrates. — Tartaric acid and the tartrates, when 
 heated in the dry state, char, and emit a very character- 
 istic odor which somewhat resembles that of burnt sugar. 
 Strong sulphuric acid blackens tartaric acid and the tartrates 
 on gently heating, while the characteristic odor of burnt 
 sugar is emitted. This is the only class of salts amon^ 
 all those within the scope of this treatise, which exhibit 
 this carbonization by sulphuric acid. 
 
 To confirm- the presence of tartaric acid, or a tartrate, 
 in any liquid supposed to contain it, a concentrated solution 
 of acetate of potassium is added to the liquid, and the mix- 
 ture violently shaken. The precipitate, when one forms, is 
 a difficultly soluble acid tartrate of potassium. The addition 
 of an equal volume of alcohol increases the delicacy of the 
 reaction. The more concentrated the solution to be tested, 
 the better. To prepare the required solution of acetate of 
 potassium at the moment of use, rub together in a dish half ^ 
 a teaspoonful of carbonate of potassium and as many drops 
 of acetic acid as will dissolve three quarters of the carbon- 
 ate ; throw the mixture on a small moistened filter and use 
 the filtrate. 
 
 Tartaric acid is a good substance from which to get these 
 reactions. 
 
96 BORATES. — SILICATES. — FLUOBIDES. [§§ 69-71. 
 
 69. Borates. — To confirm the presence of a borate, strong 
 sulphuric acid is mixed with the dry substance under exam- 
 ination in quantity sufficient to make a thin paste, and an 
 equal bulk of alcohol is added to the mixture. The alcohol 
 is then kindled. Boracic acid imparts to the alcohol flame 
 a y ellowish-green color . The test is made more delicate by 
 stirring the mixture, and by repeatedly extinguishing and 
 rekindling the flame. Copper salts impart a somewhat 
 similar color to the flame; but this metal, if present, may 
 be got rid of by sulphuretted hydrogen before testing for 
 boracic acid. 
 
 The reactions of borates may be obtained with a fragment 
 of borax. 
 
 70. Silicates. — The silicates of sodium and potassium 
 are the only silicates which are soluble in water. The 
 solutions of these alkaline silicates are decomposed by all 
 acids. If hydrochloric acid is added gradually to a strong 
 solution of an alkaline silicate, the greater part of the 
 silicic acid separates as a gelatinous hydrate. As a rule, 
 the more dilute the fluid, the more silicic acid remains in 
 solution. 
 
 If the solution of an alkaline silicate, mixed with hydro- 
 chloric or nitric acid in excess, be evaporated to dryness, 
 silicic acid separates ; if the dry mass be ignited and then 
 treated with dilute hydrochloric or nitric acid, the whole 
 of the silicic acid remains insoluble in the free state as a 
 gritty, whitish powder, while the other substances dissolve. 
 
 A solution of c hloride of ammonium produces a gelati- 
 npus prec i pitate in strong and moderately dilute solutions of 
 the alkaline silicates. This precipitate is hydrated silicic 
 acid containing alkali. 
 
 A solution of water-glass is the best substance in which 
 to study the reactions of the silicates of the alkali-metals. 
 
 71. Fluorides. — When a fluoride, naturally combined or 
 artificially mixed with silica, is heated with strong sul- 
 
§ 71.] FLUORIDES. 97 
 
 phuric acid, fluoride of silicon is evolved. This reaction 
 is available as a test for fluorine.. 
 
 A mixture of the supposed fluoride and fine dry sand is 
 heated in a short, dry test-tube with concentrated sulphuric 
 acid. A drop of water, caught in the loop of a clean plati- 
 num wire, is held in the mouth of the test-tube. This drop 
 of water becomes merely dim, quite opaque, or almost solid 
 with silicic acid, according to the quantity of fluoride of 
 silicon evolved from the mixture. The gaseous fluoride of 
 silicon shows white fumes when it comes in contact with 
 •^moist air. If a considerable quantity of fluoride of silicon 
 be evolved from the mixture tested, it can be decanted into 
 another test-tube, and there shaken up with water. If the 
 substance to be tested for fluorine is known to contain 
 silica, it is, of course, unnecessary to add sand to it. This 
 method applies to all fluorides decomposable by hot sul- 
 phuric acid. It is evident that this test reversed can be 
 applied to the detection of silica. 
 
 If a finely pulverized fluoride is heated with concentrated 
 sulphuric acid in a small leaden capsule or platinum cruci- 
 ble, hydrofluoric acid is disengaged. 
 
 Coat with wax the convex face of a watch-glass large 
 enough to cover the capsule, by heating the glass cautiously, 
 and spreading a small bit of wax evenly over it while the 
 glass is hot. Trace some lines or letters through the wax 
 with a pointed instrument of wood or horn. Fill the hollow 
 of the glass with cold water, and cover with it the capsule 
 which contains the fluoride mixture. Heat the capsule gently 
 for half an hour or an hour. Then remove the watch-glass, 
 dry it, heat it cautiously to melt the wax, and wipe it with 
 a bit of paper. The lines or letters traced through the wax 
 will be found etched into the glass. A barely perceptible 
 etching is made more visible by breathing upon the glass. 
 If much silicic acid is present, this reaction fails. 
 
 Fluoride of calcium (fluor-spar) is a good material from 
 which to obtain these two tests for fluorine. 
 
98 CHLORIDES. — BROMIDES. [§§ 72, 73. 
 
 72. Chlorides. — The following confirmatory test may be 
 applied to chlorides in the dry state. 
 
 When a c hloride , in powder, is he ated in a test-tube with 
 bl ack oxide of manganes e and strong sulphuric ^^^(^^ chlorme 
 gas is evolved; this gas is recognized by its odor, greenish- 
 yellow color and reaction with iodo-starch paper. The gas 
 evolved by a chloride gives' no colored reaction with starch 
 alone; but when a moistened slip of paper, on which a 
 mixture of starch paste and iodide of potassium (App., § 39) 
 has been spread, is held in an atmosphere or current of 
 chlorine, the paper is colored blue in consequence of the 
 lib eration of iod ine which the chlorine effects. The yellow 
 color of the gas is IBest seen by looking lengthwise through 
 the tube. 
 
 Unless it has been prepared artificially it is often diffi- 
 cult to obtain black oxide of manganese which will not of 
 itself give a reaction for chlorine when treated with pure 
 sulphuric acid. 
 
 Mercurous chloride (calomel) gives no reaction for chlo- 
 rine when treated with binoxide of manganese and sulphuric 
 acid. The chlorine may be detected, however, by treating 
 the salt in powder with zinc and dilute sulphuric acid. 
 The mercury is reduced to the metallic state, and the chlo- 
 rine goes into solution as chloride of zinc; in this solution 
 the chlorine may be discovered by the ordinary tests. 
 
 73. Bromides. — The confirmatory tests for bromides 
 depend upon the setting free of bromine itself. 
 
 Hot nitric acid liberates the bromine from all bromides 
 except those of silver and mercury. In solutions, the free 
 bromine produces a yellow coloration; when set free from 
 solid bromides in a long and narrow tube, the brownish- 
 yellow vapors of bromine condense into a liquid upon the 
 cold walls of the tube. 
 
 When bromides, in powder, are heated in a test-tube with 
 black oxide of manganese and strong sulphuric acid, 
 
§§ 73, 74.] IODIDES. 99 
 
 brownish-red vapors of bromine are evolved. If chlorides 
 are also present, the bromine will be mixed with chlorine. 
 
 To identify bromine and distinguish it from chlorine, 
 moistened starch is brought into contact with the free 
 bromine. A yellow or orange-yellow coloration of the 
 starch marks the presence of bromine. To apply this test, 
 thrust a rod smeared with starch-paste into the tube which 
 contains the bromine vapors ; or, when greater delicacy is 
 requisite, perform the experiment which is expected to 
 liberate bromine in a very small beaker, and cover this 
 beaker with a watch-glass to whose under side is attached 
 a bit of paper moistened with starch-paste and sprinkled 
 with dry starch. , 
 
 Bromide of potassium is a good substance with which 
 to study the tests for bromine. 
 
 74. Iodides. — When an iodide, in the solid form or in 
 solution, is heated with strong nitric acid, iodine is liber- 
 ated and sublimes in violet vapors. 
 
 Free iodine in vapor is recognized by the deep blue color 
 which it imparts to starch-paste. Vapors may be tested 
 by bringing into contact with them a glass rod smeared 
 with thin starch-paste, or a slip of white paper on which 
 the paste has been spread. 
 
 The best method of detecting iodftie in a solution is to 
 add a few drops of thin, clear starch-paste to the liquid, 
 and then set free the iodine by means of nitrite of potassium 
 (App., §38), as follows: — The cold fluid to be tested is 
 acidulated with dilute hydrochloric or sulphuric acid, after 
 the addition of the starch-paste, and a drop or two of a con- 
 centrated solution of nitrite of potassium is then added. 
 A dark blue color will be instantly produced. It is essen- 
 tial that the liquid should be kept cool, for the blue colora- 
 tion is destroyed by heat. 
 
 Like chlorine and bromine, iodine is liberated by heating 
 an iodide with black oxide of manganese and sulphuric acid. 
 
100 IODIDES. — BROMIDES. [§ 74. 
 
 The iodine so liberated is readily distinguished by the 
 above tests. 
 
 The student can try all these tests for iodine with a small 
 crystal of iodide of potassium. 
 
 The tests given in §§ 72-74 are not always satisfactory or 
 conclusive. The following method is preferable and at 
 the same time enables the analyst to test for those four 
 acids forming insoluble salts of silver successively and in 
 the presence of each other in a satisfactory manner. 
 
 In case a nitric acid solution or aqueous solution acidi- 
 fied with nitric acid of any substance gives a white or 
 whitish precipitate on the addition of nitrate of silver solu- 
 tion, chlorides, bromides, iodides or cyanides are presum- 
 ably present. In ordinary cases a mere turbidity indicates 
 the presence of a chloride, and other acids of the group may 
 be regarded as absent. In case a white or whitish precipi- 
 tate insoluble in nitric acid is produced on the addition of 
 nitrate of silver, proceed for the detection of the acid as 
 follows : — 
 
 a. Boil about a. half a gram of the original substance with 
 two or three teaspoonfuls of a saturated solution of carbonate 
 of sodium for about ten minutes, replacing the water that 
 evaporates ; dilute with an equal volume of water, and filter. 
 
 b. Iodides. To a portion of this filtrate add nitric acid 
 to strongly acid reaction and a few drops of chromate of 
 potassium solution, then add a few drops of carbon bisul- 
 phide (App., § 64). Shake thoroughly; violet coloration of 
 the carbon bisulphide shows the presence of iodine. 
 
 c. Bromides. If no violet coloration of the bisulphide of 
 carbon appears, add a few drops of chlorine water (App., 
 § 65), to the same test, and shake thoroughly; brown col- 
 oration of the bisulphide of carbon shows the presence of 
 bromine. 
 
 In case iodine was found in the preceding test, filter out 
 the violet-colored bisulphide of carbon obtained through a 
 
§ 74.] CYANIDES. — CHLORIDES. 101 
 
 wet filter, add half a teaspoonful of chlorine water to the 
 filtrate, shake thoroughly, add a few drops of bisulphide of 
 carbon, and shake again ; bromine, if present, will color the 
 bisulphide of carbon brown. 
 
 d. Cyanides. Acidify a portion of the original filtrate 
 with dilute sulphuric acid, and test as described in the test 
 for cyanides previously given (§ 59). 
 
 e. Chlorides. If no reaction is obtained for iodine, 
 bromine or cyanogen in the above tests, the white precipitate 
 produced by the nitrate of silver can only be due to the 
 presence of a chloride, and no further test need be made. 
 
 In case iodine, bromine or cyanogen has been found, 
 acidify a portion of the original filtrate (a) with nitric acid, 
 add nitrate of silver solution to complete precipitation, 
 wash several times by decantation and proceed as follows : — 
 
 /. Iodine only was found. Warm the precipitate gently 
 for a few minutes with ammonia-water, filter, and acidify 
 the filtrate with nitric acid: chloride of silver is precipi- 
 tated if present. 
 
 g. Bromine was found, or bromine and iodine; cyanogen 
 being absent. Add to the precipitate produced by nitrate of 
 silver a teaspoonful of carbonate of ammonium solution, 
 pass carbon dioxide through the mixture for a short time, 
 warming gently; filter, and acidify the filtrate with nitric 
 acid : a white precipitate shows a chloride. 
 
 h. Cyanogen is present. Dry in a small porcelain cruci- 
 ble or evaporating-dish, at a gentle heat, the precipitate pro- 
 duced by addition of nitrate of silver to the original filtrate 
 acidified with nitric acid, ignite to low redness and allow to 
 cool. Place a bit of zinc in contact with the residue, add 
 a little dilute sulphuric acid, and allow action to continue 
 for an hour or more ; dilute with an equal volume of water, 
 filter, and add to the clear liquid nitrate of silver solution. 
 The precipitate thus obtained may be tested as directed in 
 a, b, c and d; but in the absence of bromides or iodides a 
 
102 NITRATES. [§§ 74, 75, 
 
 white precipitate at this point is a sufficient test for a 
 chloride. 
 
 75. Nitrates. — The preliminary examination generally 
 gives warning of the presence of a nitrate : to confirm the 
 presence of a nitrate, one or both of the two following tests 
 may be used : — 
 
 If the solution of a nitrate is mixed with an equal volume 
 of s trong sulphuric acid, the mixture thoroughly cooled in 
 cold water, and a concentrated solution of fe rrous sulp hate 
 then cautiously added to it in such a way that the two fluids 
 do not mix, the stratum of ^contact shows a purple or red- 
 dishcolor, which changes to a brown. If the fluids are then 
 mixed, a clear brownish liquid is obtained. The color fades 
 on heating. Another way of performing the same test is 
 to drop a crystal of ferrous sulphate into the cold mixture of 
 nitrate and sulphuric acid. There forms around the crystal 
 a dark halo, which disappears with a kind of effervescence 
 on the application of heat. It is, of course, essential that 
 the sulphuric acid employed for this test should be so free 
 from nitric and hyponitric acids, as not itself to give this 
 reaction with ferrous sulphate. Chromic acid interferes 
 with this. test, and if a chromate is present, it must be 
 removed by precipitation with acetate of lead. 
 
 Boil some hydrochloric acid in a test-tube, add to it one 
 or two drops of a dilute solution of sulphindigotic acid 
 (App., § 58), and continue the boiling a moment. If the 
 chlorhydric acid is sufficiently free from chlorine, the 
 resulting liquid will be of a faint blue color. If a nitrate, 
 either solid or in solution, be added to this liquid and the 
 mixture be again boiled, the liquid will be decolorized. 
 This reaction is delicate; but there are some other sub- 
 stances, especially free chlorine, which have a like bleach- 
 ing effect. 
 
 Nitrate of potassium is a suitable material on which to 
 illustrate the tests for nitrates. 
 
§§ 76, 77.] CHLORATES. — ACETATES. 103 
 
 76. Chlorates. — The preliminary examination gives warn- 
 ing of the presence of chlorates. 
 
 When a few particles of a chlorate in the solid form are 
 covered with two or three times as much strong sulphuric 
 acid , and the mixture is g ently warme d, the liquid becomes 
 intensely yellow, and a greenish-yellow irritating gas of 
 peculiar odor (chlorine dioxide, CIO2) is evolved, which ex- 
 plodes with violence at a moderate heat. After this decom- 
 position, the gas evolved has the characteristic odor of 
 chlorine. The quantity of chlorate operated upon sht)uld 
 be very small. 
 
 The solution of a chlorate decolorizes indigo-solution 
 precisely like the solution of a nitrate, under like condi- 
 tions (§ 75). 
 
 A chlorate is converted by ignition into a chloride, from 
 a solution of which nitrate of silver precipitates the chlo- 
 rine. 
 
 Chlorate of potassium illustrates very well the reactions 
 of chlorates. 
 
 T*" Acetates. — When acetates are moderately heated 
 with strong sulphuric acid, hydrated acetic acid distils from 
 the mixture, and may be recognized by its pungent odor. 
 
 When an acetate is heated with alcohol and sulphuric 
 acid in equal volumes, acetic ether is formed. The agree- 
 able odor of this ether is highly characteristic. To be sure 
 of this odor, and not mistake it for that of other ethers, a 
 comparison test should always be made with a known ace- 
 tate at the same time. 
 
 Hot concentrated sulphuric acid produces no blackening 
 with an acetate. 
 
 When a few drops of a solution of ferric chloride (App., 
 § 49) are added to a solution of a neutral acetate, or to a 
 solution of an acetate previously neutralized with ammonia, 
 the liquid acquires a dark red color, because of the forma- 
 tion of ferric acetate. If the liquid contain an excess of 
 
104 OXIDES AND HYDRATES. 
 
 the acetate, a basic acetate of iron is precipitated in yellow 
 flocks upon boiling, and the fluid finally becomes colorless. 
 
 Acetate of sodium may be used to show these reactions. 
 
 78. Oxides and Hydrates. — In case the substance under 
 examination is an oxide or a hydrate, it may in general 
 be recognized by the known properties of the oxide or 
 hydrate of the metallic element which has been found to 
 be present. The hydrates (except of the alkalies) give 
 off water when heated, and many oxides and hydrates when 
 he^ed with charcoal in a closed tube give off carbonic acid 
 or carbonic oxide. Peroxides may generally be distin- 
 guished by their giving off oxygen when heated, or by 
 causing an evolution of chlorine when treated with hydro- 
 chloric acid. 
 
 Of course, salts of most of the oxy-acids, and, indeed, 
 many other substances besides oxides and hydrates, give 
 carbonic acid when heated with carbon. This test miay 
 serve to identify a simple substance, but the presence or 
 absence of an oxide in a mixture of salts can often be deter- 
 mined only by quantitative analysis. 
 
Part Second. 
 
 THE ACTUAL EXAMINATION OP SUBSTANCES OF UN- 
 KNOWN COMPOSITION. 
 
 79. The substance to be examined may be either solid 
 or liquid. We shall consider first the treatment of a solid; 
 afterwards that of a liquid. The solid may be a metallic 
 substance, that is, a pure metal or an alloy, or it may be a 
 salt, mineral or other non-metallic body. The method of 
 procedure differs in the two cases. We shall describe first 
 the treatment of a salt, mineral or other non-metallic sub- 
 stance. The two following observations, however, apply 
 to all cases. The student should, in the first place, learn as 
 much as possible from the external properties of the sub- 
 stance to be analyzed, from its color, consistency and odor, 
 if it is a liquid; from its color, texture, odor, lustre, hard- 
 ness, gravity and crystalline or amorphous structure, if it 
 is a solid. By attentively observing the characteristics or 
 individual peculiarities of every substance which passes 
 through his hands, the student will soon learn to recognize 
 many substances at sight, — by far the quickest and easiest 
 way of identifying them. Secondly, since the original sub- 
 stance must be several times reverted to in order to com- 
 plete an analysis, the student should husband his stock of 
 the substance to be analyzed, never employing the whole of 
 it for any single course of experiment. It is well also to 
 reserve a portion for unforeseen contingencies. 
 
 105 
 
CHAPTER XI. 
 
 TREATMENT OF A SALT, MINERAL OR OTHER NON- 
 METALLIC SOLID. 
 
 Order of Procedure. 
 
 80. The general course pursued in the analysis of a 
 non-metallic solid substance consists, as a rule, in: — 1st. 
 The preliminary examination, to be presently described; 
 2d. The bringing the substance into solution; 3d. The 
 examination of the solution obtained for the metallic ele- 
 ments according to the methods laid down in Part I, Chaps. 
 I-VIII; 4th. The application of such general and special 
 tests for the non-metallic elements as may not have been 
 rendered unnecessary by facts observed during the preced- 
 ing course of the analysis. 
 
 A. Preliminary Examination in the Dry Way. 
 
 81. The preliminary examination in the dry way consists 
 essentially of two operations; the substance is first sub- 
 jected to the influence of heat alone (closed-tube test, § 82) ; 
 and afterwards a second portion of the substance, or, in 
 some cases, the same portion, is exposed to the influence of 
 heat and certain reducing agents (reduction test, § 83). 
 
 82. Closed-Tube Test. — Prepare a hard glass tube No. 5 
 (App., § 86), about three inches long, and closed at one 
 end. Let fall into this tube a minute fragment of the 
 solid, or a little of its powder. If the substance be used in 
 powder, wipe out the tube with a tuft of cotton on a wire, 
 in order that the interior walls of the tube may be clean to 
 receive a sublimate. Heat the substance at the end of the 
 
 106 
 
§ 82.] CLOSED-TUBE TEST. 107 
 
 tube, at first gently in the lamp, but finally intensely at 
 the highest temperature to be obtained with the Bunsen 
 gas -lamp, or in the blowpipe flame. The following are the 
 most noteworthy reactions, with the inferences to be drawn 
 from them ; it not unfrequently happens that a single sub- 
 stance gives several of these reactions. 
 
 I. a. The substance blackens, and gases or vapors are 
 evolved. These vapors often have a disagreeable smell, 
 sometimes like that of burnt sugar, paper or feathers. 
 Sometimes they condense in tarry droplets ; water also con- 
 denses on the cold part of the tube. These appearances 
 indicate the presence of organic substances. 
 
 Now the presence of fixed organic matter interferes with 
 the detection of many substances, and it must, as a rule, 
 be destroyed before the analysis can be proceeded with. 
 The method employed is detailed on page 115; no matter 
 whether the substance contains organic matter or not, the 
 closed-tube test is immediately succeeded by the reduction 
 test. Simple blackening is not proof of the presence of 
 organic bodies. Some salts of copper and cobalt, for exam- 
 ple, blacken through the formation of a black oxide. 
 
 When organic matter is shown to be present, the student 
 should look particularly, at the proper stage of the analysis, 
 for acetates (§ 77) and tartrates (§ 68); but he will not 
 forget that there are hundreds of organic compounds which 
 are not comprehended in the plan of this treatise. 
 
 b. Certain other changes of color are important. Zinc 
 oxide and most salts of zinc are yellow when hot, but 
 become white on cooling. Lead oxide and many lead salts 
 are yellow while hot, and remain yellow on cooling. Bis- 
 muth oxide and many bismuth salts are an orange to red- 
 brown while hot, pale yellow on cooling. Ferric oxide and 
 ferric salts are red to black while hot, reddish-brown when 
 cold. Oxide of tin, yellowish on heating, white when cold. 
 Red oxide of mercury, black when hot, red again on cooling. 
 
108 CLOSED-TUBE TEST. [§ 82. 
 
 II. The substance is not carbonized, but vapors or gases 
 escape from it. The most important are: — 
 
 a. Aqueous vapor, which condenses in the upper part of 
 the tube. This may indicate the presence of a salt contain- 
 ing water of crystallization, or the presence of a hydrate, or 
 of water, held mechanically. Test the water with litmus 
 paper; if it is alkaline, ammonia may be suspected; if acid, 
 some volatile acid (H,,SO„ HCl, HBr, HI, HFl, HNO3, etc.). 
 
 b. Oxygen recognized by its relighting a glowing match. 
 This gas indicates nitrates, chlorates and peroxides, or a 
 decomposable oxide (as HgO). If the heated substance 
 fuses, and a small fragment of charcoal thrown in is ener- 
 getically consumed, the presence of a nitrate or chlorate 
 may be assumed. 
 
 c. Sulphurous acid gas, recognized by its odor. It not 
 unfrequently results from the decomposition of sulphites, 
 sulphates and sulphides. 
 
 d. Carbonic acid gas, derived from decomposable carbon- 
 ates and some oxalates, recognized by causing turbidity in 
 a drop of lime-water exposed to the escaping gas. 
 
 e. Cyanogen, recognized by its peculiar odor and by the 
 violet flame with which it burns when there is enough of it 
 to be lighted, derived from decomposable cyanides. 
 
 /. Sulphuretted hydrogen, derived from moist sulphides, 
 to be known by its odor and action on lead paper. 
 
 g. Acetone, from the decomposition of acetates, recog- 
 nized by its characteristic fragrant odor. 
 
 h. Ammonia, resulting sometimes from the decomposition 
 of ammoniacal salts, recognized by its odor.V '^^<^ - '' -^^^"^^ /^uj^*- 
 
 III. Colored gases or vapors are set free. 
 
 a. Nitrogen peroxide, recognized by the brownish-red 
 color of the fumes. It results from the decomposition of 
 nitrates. 
 
 b. Bromine, recognized by its heavy brownish fumes and 
 peculiar odor, indicating bromides. 
 
§ 82.] CLOSED-TUBE TEST. 109 
 
 c. Iodine^ violet fumes and peculiar odor, indicating 
 iodides. 
 
 IV. A sublimate forms beyond the heated portion of the 
 tube. The whole of the substance may volatilize. 
 
 Some substances, dissolving in the water of crystalliza- 
 tion on heating, give deposits on the sides of the tube which 
 may be mistaken for true sublimates. 
 
 The following are the commonest sublimates : — 
 
 a. Ammonium salts give white sublimates as a rule. 
 Unless the original solid is obviously unalterable by heat, 
 it should be invariably tested for ammonium by heating in a 
 small test-tube with an equal bulk of slaked lime (App., 
 § 41) and a few drops of water. Ammonia, when evolved, 
 may be recognized by its smell and by the dense white 
 fumes produced when a rod, moistened with hydrochloric 
 acid, is held above the mouth of the tube. 
 
 6. Antimony trioxide first fuses to a yellow liquid, and 
 then gives a white sublimate composed of needle-like 
 crystals. 
 
 c. Oxalic acid gives a white crystalline sublimate with 
 dense fumes in the tube. 
 . d. Metallic mercury, and some of its compounds. 
 
 The metal sublimes in minute globules if present in the 
 first state, or in the form of easily reducible compounds, as 
 mercuric oxide. Mercurous chloride forms a sublimate 
 without first melting, which is yellow when hot; white 
 when cold. 
 
 Mercuric chloride melts first, then sublimes, forming a 
 white crystalline sublimate. 
 
 The sulphide of mercury gives a dull black sublimate. 
 The red iodide of mercury gives a yellow sublimate. In 
 case indications of the presence of mercury are obtained, 
 they should be confirmed by mixing a little of the thor- 
 oughly dry substance with dry carbonate of sodium, and 
 heating in a closed tube. All compounds of mercury are 
 
110 CLOSED-TUBE TEST. [§82. 
 
 thus decomposed, giving a sublimate of metallic mercury, as 
 a gray mirror on the cold part of the tube. With a lens 
 this is seen to consist of minute globules ; and by gently 
 rubbing with a wire, globules, visible to the unaided eye, 
 can be obtained. 
 
 e. Arsenic, and some of its compounds. Metallic arsenic, 
 and some arsenical compounds, give a black sublimate of 
 metallic lustre. The sulphides of arsenic give sublimates 
 which are brownish-red, sometimes nearly black when hot, 
 but reddish-yellow to red when cold ; these sublimates look 
 not unlike that of pure sulphur. Arsenious oxide gives a 
 white sublimate, seen to be crystalline by a lens. 
 
 /. Sulphur, free or by the decomposition of some sul- 
 phides, gives a sublimate of reddish-brown drops, which 
 becomes solid and yellow, or yellowish-brown, on cooling. 
 
 Some other reactions of less importance may occur in the 
 closed tube, and, in any event, it is to be observed that 
 the inferences to be drawn from the appearances already 
 indicated are of very unequal value. Thus the detection 
 of organic matter is of importance, because, as has been 
 stated, such matters must generally be got rid of before the 
 analysis can be proceeded with. Again, the presence, or 
 entire absence, of ammonium salts should be put beyond 
 doubt at this first stage of the examination. Thirdly, the 
 presence of mercury, or of mercurous salts, determines the 
 choice of the acid solvent in favor of nitric acid, in case 
 water will not dissolve the substance under examination 
 (§ 87), and the presence of mercuric salts renders necessary 
 the substitution of sulphide of ammonium for sulphide of 
 sodium as the solvent for the sulphides of Class III (§ 27). 
 Accordingly, it is useful to get information of the pres- 
 ence of mercury or its compounds at this early stage of 
 the examination. As to the other appearances, they give 
 information which may be convenient, but is never essential 
 for the safe conduct of the regular course of analysis. 
 
§§ 82, 83.] BEBUCfiON TEST. Ill 
 
 Practically the closed-tube test may generally be conducted 
 as follows : — 
 
 The substance is introduced into the tube as stated above, 
 a strip of moistened litmus paper is folded loosely across 
 the mouth of the tube, and the tube heated in the lamp, at 
 first very gently. The deportment of the substance is 
 observed under the gentle heat at first applied and also as 
 the heat increases, the appearance of condensed moisture 
 or of other sublimate is looked for, and any action of escap- 
 ing vapors on the litmus paper is noted. Meanwhile the 
 tube is occasionally removed from the lamp, to see whether 
 any peculiar odor of escaping gas may be detected. Finally, 
 any water which may have condensed in the tube is tested 
 with litmus paper. The student should note any other 
 appearances, whether mentioned above or not, which mani- 
 fest themselves during the examination, as they may serve 
 as corroborative tests when the composition of the substance 
 has finally been made out with some certainty. 
 
 83. Reduction Test. — Mix a little of the powder of the 
 substance under examination (the bulk of a hemp-seed) 
 with an equal quantity of carbonate of sodium, and make 
 the mixture into a pasty ball with a small drop of water. 
 Select a piece of dry, well-burned, soft-wood charcoal, and 
 cut out of it a rectangular block about 6 inches long, H in. 
 wide, and -J- to f in. thick, having its flat, smooth surface 
 (6 in. by 1^ in.) at right angles to the rings of growth in 
 the tree. It is this surface which is always to be used. A 
 good piece of charcoal may be made to serve for many assays 
 by filing off the used surface and exposing a new one, but 
 ordinary charcoal allows portions of the flux or even small 
 metallic globules to run into its pores or into cracks opened 
 by the heat, and can rarely be used with advantage the sec- 
 ond time. At a quarter to half an inch from the end of the 
 piece of charcoal, scoop out with a penknife a little cavity 
 of the size of half a pea. Place the prepared pellet in this 
 
112 REDUCTION TEST. [§ 83. 
 
 cavity, and expose it for several consecutive minutes to the 
 reducing-fiame of the blowpipe (App., § 82). 
 
 Under these conditions, vapors of characteristic odor or 
 appearance may be evolved ; some of them will be men- 
 tioned below. The two objects, however, to which atten- 
 tion is specially to be directed are the residue in the cavity, 
 and the incrustation on the charcoal outside of the cavity. 
 
 The following metals may be found as fused metallic 
 globules in the cavity; lead, silver and gold are reduced 
 with ease, even by an inexperienced operator ; tin and copper 
 with some difficulty : — 
 
 a. Gold — a yellow, malleable globule, produced without 
 incrustation. 
 
 b. Copper — a red, malleable globule, produced without 
 incrustation. 
 
 c. Tin — a bright, white, malleable globule. An incrus- 
 tation is simultaneously produced, which is faint yellow 
 when hot and white when cold; it immediately surrounds 
 the globule. 
 
 d. Lead — a very fusible and very malleable globule. 
 A yellow incrustation is simultaneously produced. 
 
 e. Silver — a brilliant, white, malleable globule, produced 
 without incrustation. 
 
 /. Bismuth — a dark orange-yellow incrustation, chang- 
 ing to lemon-yellow on cooling. A gray, brittle, metallic 
 globule. 
 
 g. Antimony — a gray metallic globule. A white in- 
 crustation. 
 
 h. Cadmium — a reddish-brown incrustation; in thin lay- 
 ers orange-yellow. 
 
 Common charcoal is itself very apt to show a grayish 
 incrustation of ash round about the heated assay; this in- 
 crustation remains unaltered or increases, when directly 
 exposed to the flame. The student should test each piece 
 of charcoal before the blowpipe flame^ in order that he may 
 
§ 83.] nSDXTCTlON TEST. 113 
 
 not imagine a deposit of ash to be an incrustation derived 
 from the substance under examination. 
 
 If a distinct globule has been obtained, it must be picked 
 out with a pair of jewellers' tweezers, and pounded on some 
 smooth and hard body to test its malleability. If it is mal- 
 leable, replace it upon the charcoal at an unused spot, and 
 lieat it strongly with the oxidizing flame. Gold and silver 
 globules fuse, but maintain their brilliancy and give no 
 incrustation ; this proof distinguishes a genuine gold globule 
 from a yellow globule composed of an alloy of copper and 
 some white metal. A yellow globule composed of oxidiz- 
 able metals tarnishes instantly in the oxidizing flame. A 
 tin globule fuses, but its bright surface is instantly tar- 
 nished, and a white incrustation of binoxide of tin is pro- 
 duced which cannot be driven off by either flame. A lead 
 globule is rapidly converted into litharge, a yellow incrus- 
 tation being produced, which volatilizes with a bluish 
 color when touched with the reducing-flame. A copper 
 globule is blackened by the formation of oxide of copper, 
 and the blowpipe flame is tinged with green. 
 
 Certain other phenomena may manifest themselves dur- 
 ing this experiment for the reduction of malleable metallic 
 globules. Sulphur, ammonium salts in general, the chlo- 
 rides, bromides, iodides and sulphides of sodium and potas- 
 sium, the chlorides of lead, bismuth, tin and copper, metallic 
 mercury, arsenic, antimony and zinc, and many compounds 
 of these four elements, are liable to pass off in vapors, 
 Avhich are often in part deposited upon the coal at a greater 
 or less distance from the hot assay, according to their vol- 
 atility. With the exception of sulphur, these sublimates 
 are white, but when deposited in a thin film upon the black 
 coal they have a gray or blue appearance. During the pro- 
 duction of the arsenic sublimate a peculiar odor is evolved; 
 this sublimate being very volatile is only deposited at a con- 
 siderable distance from the assay. The incrustation pro- 
 
114 HtltotlCTlON TEST. [§ 83. 
 
 duced by zinc is distinctly yellow while hot, but turns white 
 on cooling; it settles near the assay and is driven away 
 again with difficulty. Nitrates and chlorates generally give 
 warning of their presence when heated on charcoal by caus- 
 ing deflagration. 
 
 The main object of the experiment is the reduction of the 
 five malleable metals above enumerated. We may thus 
 obtain knowledge of the presence of gold (for a confirma- 
 tory test, see § 98, b). Tin generally gives warning of its 
 presence during this experiment ; and this warning is of use, 
 because it is inconvenient to apply nitric acid as a solvent 
 to a substance containing tin, since this reagent converts tin 
 into the very insoluble binoxide of tin. The detection of 
 copper at this stage is of little advantage. The most im- 
 portant fact deducible from the reduction test is the pres- 
 ence of either silver or lead. In dissolving an unknown sub- 
 tance which proves to be insoluble in water, it is customary 
 to try, as the second solvent, hydrochloric acid. The chlo- 
 rides of silver and lead being insoluble, or difficultly soluble, 
 this acid should not be used as a solvent when either of 
 these two metals is present. Nitric acid must be used in- 
 stead. Moreover, if the substance has already given evi- 
 dence of the presence of organic matter (§ 82), porcelain, 
 and not platinum, must be used as a support in destroying 
 the organic matter, as described in the next section, when- 
 ever an easily reducible metal is present. 
 
 Sometimes, when the substance under examination con- 
 tains but a small proportion of metal, some metal may be 
 reduced during the blowpipe experiment on charcoal, but 
 the detached particles may not run together into a single 
 conspicuous globule. Since a mistake as to the presence of 
 a reducible metal may involve the destruction of a plati- 
 num crucible, it is best in doubtful cases to operate in a 
 more delicate fashion. To ascertain, beyond question, 
 whether any reduced metal has been separated in this 
 
§§88,84.] DESTRUCTION OF ORGANIC MATTER. 115 
 
 experiment, moisten the cavity in the charcoal with water 
 after the fusion has been finished, cut the charcoal out for 
 a little distance, both around and below the cavity, and 
 transfer the contents of the cavity and the scraps of charcoal 
 to an agate or porcelain mortar. Pulverize the whole mass, 
 and then carefully wash away the powdered charcoal and all 
 the lighter portion of the mixture. Any malleable metal 
 that may have been reduced remains in the mortar in little 
 flattened grains or spangles, in which the peculiar color and 
 lustre of the metal or alloy are generally visible. Sometimes 
 metallic streaks are produced on the mortar or pestle by 
 little particles of metal ground between them. 
 
 Iron, nickel and cobalt when reduced on charcoal remain 
 aftei? this washing as black or dark-gray powders, which are 
 attracted by a magnet. 
 
 The student must not mistake glistening particles of wet 
 charcoal sticking to the mortar or pestle for metallic span- 
 gles, and the metal should be thoroughly removed from both 
 mortar and pestle by the use of a few drops of warm aqua 
 regia, immediately after the experiment is finished, in order 
 to avoid errors on a subsequent occasion. 
 
 84. Destruction of the Organic Matter. — As already 
 stated in § 82, the presence of fixed organic matter interferes 
 with the detection of many substances, and it must, as a 
 rule, be destroyed before the analysis can proceed. A por- 
 tion of the original substance, sufficient for the regular 
 course of examination for the metallic elements, is ignited 
 in a porcelain crucible, with free access of air, or better, on 
 platinum foil, if the absence of any easily reducible metal 
 has been proved by the reduction test, until all the carbon is 
 burnt out of it. This ignition is best performed on succes- 
 sive small portions rather than on a large mass at once. 
 
 It is obvious that some inorganic volatile matters may be 
 lost during this ignition. Furthermore, some substances, 
 especially alumina and chromic and ferric oxides, are made 
 
116 SOLUTION. [§§84, 8& 
 
 very insoluble by ignition. Exceptionally, therefore, the fol- 
 lowing process, which is not liable to these objections, is 
 employed : The substance in powder, paste or concentrated 
 solution, is heated in an evaporating-dish, with strong nitric 
 acid, to a temperature just below boiling. To this hot mix- 
 ture chlorate of potassium, in small bits, is added gradually 
 during several minutes, or until the organic matter is all 
 destroyed. The solution is then evaporated to dryness on a 
 water-bath; the dry residue is moistened with strong hydro- 
 chloric acid, the mixture diluted with water, warmed and 
 filtered, if there be any residue. The filtrate is fit for the 
 regular course of analysis, except that potassium, having 
 been added, must not be tested for in this liquid. The 
 residue, if any, must be examined for the insoluble chlorides 
 of Class I. This process is simply a combustion at a low 
 temperature. 
 
 The only classes of salts, coming within the scope of 
 this manual, which in the closed-tube give evidence of the 
 presence of organic matter, are the acetates and tartrates. 
 
 It will, of course, be necessary to apply the tests for acetic 
 and tartaric acid to a portion of the original substance and 
 not to the solution obtained above. But the student will 
 not forget the statement already made in § 82, that the 
 presence of organic matter does not necessarily imply the 
 presence of either acetates or tartrates. 
 
 B. Dissolving a Salt, Mineral or other Non-Metallic Solid, 
 free from Organic Matter. 
 
 85. Before a solid substance can be submitted to the 
 systematic course of analysis, it must be brought into solu- 
 tion. There is no universal solvent. Different substances 
 require different solvents. The four solvents employed in 
 qualitative analysis for salts, minerals and other non-metal- 
 lic solids, are water, hydrochloric acid, nitric acid and aqua 
 
§§ 85, 86.] SOLUTION. 117 
 
 regia; and these four liquids are to be tried in the precise 
 order 'in which they here stand. Water is always to be tried 
 first; to whatever resists water, strong hydrochloric acid is 
 applied ; if hydrochloric acid fails to dissolve the solid com- 
 pletely, nitric acid is tried, and after nitric acid, aqua regia, 
 as the last resort. If, for reasons stated in § 83, the pre- 
 liminary examination has shown hydrochloric acid to be 
 unsuitable as a solvent, nitric acid is tried immediately after 
 water, and lastly aqua regia as before. A solid substance 
 should invariably be reduced to a very fine powder before 
 being submitted to the action of solvents (App., § 95). 
 
 86. Dissolving in Water. — About half a thimbleful of 
 the powdered substance is boiled with ten times as much 
 water in a test-tube. If an effervescence occurs, as is 
 possible with mixtures containing an acid salt (yeast-pow- 
 ders, for example), the gas evolved should be carefully 
 tested (§§ 57-62). In attempting to dissolve a substance it 
 should be remembered that several salts, whilst readily dis- 
 solving in a small amount of water, are decomposed by large 
 amounts and rendered insoluble. The addition of a little free 
 acid will usually prevent this decomposition or redissolve 
 the basic salt formed. Several substances also, which are 
 soluble in cold or warm water, undergo decomposition on 
 boiling. On this point consult § 88. If the substance dis- 
 solves completely, the solution is ready for analysis. When 
 undissolved powder remains in the tube after protracted 
 boiling, filter a few drops of the liquid, and evaporate a drop 
 or two of the filtrate to dryness on clean platinum foil, at 
 as low a heat as possible. If there be no residue on the 
 foil, or if the residue be scarcely appreciable, the substance 
 is practically insoluble in water, and acids must be tried as 
 solvents. But if, on the contrary, a tolerable residue re- 
 mains on the foil, decant the liquid in the tube into the 
 filter, and boil the powder again with water. Persevere 
 with this treatment until it is evident that a part of the 
 
118 SOLUTION, [§§86,87. 
 
 powder is insoluble in water. The insoluble residue in the 
 test-tube is hltered off; the clear filtrates, collected together 
 and concentrated by evaporation, if of unreasonable bulk, 
 are to be labelled "H2O Sol." and reserved for the regular 
 course of analysis (§ 88). In this case, and in the still 
 more favorable case in which all the substance has dissolved 
 in water, it is a simple aqueous solution which is submitted 
 to analysis. 
 
 87. Dissolving in Acids. — The substance which water has 
 failed to dissolve, either in whole or in part, is next boiled 
 in a small dish with three or four times its bulk of concen- 
 trated hydrochloric acid, unless the tube test (§ 82 ) or the 
 reduction test (§ 83) has proved the presence of silver, lead 
 or mercury; in which case nitric acid is the first acid to be 
 tried. If an effervescence occur, the escaping gas is to be 
 tested, as described in §§ 57-62. (See also the last part 
 of § 78.) After boiling the powdered substance with the 
 strong acid, dilute the fluid with twice its bulk of water, and 
 repeat the boiling if any residue remain undissolved. The 
 acid is diluted because, though the substance to be dissolved 
 is best attacked in the first instance by strong acid, the salts 
 formed by the action of the concentrated acid are more 
 likely to dissolve readily in a dilute than in a strongly acid 
 liquor. Not a few salts, which scarcely dissolve in strong 
 acids, are readily soluble in the same acids when diluted. 
 If the whole of the substance finally dissolves, the solution, 
 still farther diluted, is ready for the transmission of sulphu- 
 retted hydrogen (§ 20), for it is of course unnecessary to 
 examine it for members of Class I. If, on the contrary, an 
 undissolved residue remain in the tube, ascertain if any- 
 thing has dissolved, by carefully evaporating two or three 
 drops of the fluid to dryness on platinum foil. Should an 
 appreciable residue, in excess of that given by two or three 
 drops of the acid employed, remain upon the foil, separate 
 the liquid in the tube from the undissolved substance by 
 
 Ik 
 
§ 87.] SOLUTION. 119 
 
 decantation or filtration. Reserve the solution, labelling 
 it "HClSol." 
 
 Rinse the undissolved powder with water, and then boil 
 it in an evaporating-dish with three or four times its bulk 
 of strong nitric acid. If the original substance contains 
 silver, lead or a mercurous salt, hydrochloric acid will not 
 have been used, and it will be the residue from the aqueous 
 solution, which is now to be boiled with nitric acid. In this 
 case effervescence is to be watched for. If the substance 
 dissolves completely in the strong acid, or dissolves, with 
 the exception of a light yellow mass of sulphur, which often 
 separates from a sulphide, evaporate the liquid almost to dry- 
 ness to drive olf the free acid, dilute the evaporated solution 
 with several times its bulk of water, separate the sulphur, 
 if necessary, by filtration, and reserve the solution, label- 
 ling it "HNO3 Sol." If the substance does not completely 
 dissolve in the strong acid, dilute the fluid with twice its 
 bulk of water, and repeat the boiling. If the dilute nitric 
 acid effects complete solution, reserve the solution, labelling 
 it as before, "HNO3 Sol." If neither the strong nor the 
 diluted nitric acid effects the complete solution of the sub- 
 stance, ascertain if anything has dissolved in the dilute 
 acid by the usual test on platinum foil. If an appreciable 
 residue remain on the foil, separate the undissolved solid 
 in the dish from the liquid by decantation or filtration, and 
 reserve the solution. 
 
 Boil the powder, which has resisted both acids taken 
 singly, with aqua regia. If it dissolves completely, evapo- 
 rate the solution almost to dryness, redissolve in water, 
 with the addition of a little strong hydrochloric acid. If 
 there be any turbidity, dilute largely with water and reserve 
 the solution for analysis, labelling it, "Aq. Reg. Sol." It 
 is useless to look for the members of Class I in such a 
 solution. If, on the contrary, it does not completely dis- 
 solve after protracted boiling, test the liquor to see if any- 
 
120 AN AQUEOUS SOLUTION. [§§ 87, 88. 
 
 thing has dissolved. If an appreciable residue remains on 
 the foil, dilute the acid fluid, filter it, reserve the solution 
 labelled as before, and wash the undissolved residue thor- 
 oughly with water, to prepare it for further treatment (§ 88). 
 
 Certain silicates, when boiled with concentrated acids, are 
 decomposed, and gelatinous silicic acid is separated. This 
 happens but rarely, however, in the rapid processes of quali- 
 tative analysis; and if it should happen, it is not likely to 
 lead the student into error. A residue insoluble in all acids 
 will remain; this residue is, or contains, free silicic acid. 
 
 It must not be supposed that it is common to try all four 
 solvents on one and the same substance. Water and hy- 
 drochloric acid are the common solvents; nitric acid and 
 aqua regia are, actually, but seldom resorted to as solvents, 
 except for metals (§98). It would require some ingenuity 
 to devise an artificial mixture which would put to the test 
 all the capabilities of the above-described method of bring- 
 ing solids into solution in water and acids. Such mixtures 
 are not met with in ordinary experience. It is the object 
 of any method of analysis to meet real problems, not 
 artificial complications which may be imagined but which 
 do not occur in fact. 
 
 C. Treatment of the Solutions Obtained. 
 
 88. Since substances which prove to be insoluble in water 
 and in acids must be brought into solution by peculiar 
 methods, we proceed to consider first the treatment of the 
 solutions already obtained. 
 
 I. An Aqueous Solution. — If the student is assured that 
 the unknown substance is a simple salt, he may draw some 
 trustworthy inferences from the behavior of the substance 
 when solution in water is attempted. Of the salts which 
 fall within the scope of this manual, the following general 
 statements may be made : — 
 
§ 88.] AN AQUEOUS SOLUTION. 121 
 
 1. All salts of sodium and all of potassium and ammo-, 
 nium, except their double platinum-chlorides, are practically 
 soluble in water. 
 
 2. All normal nitrates, chlorates and acetates are soluble. 
 But some of the basic salts of these acids are insoluble. 
 The nitrates of mercury undergo decomposition on boiling 
 with water, insoluble basic compounds being formed. Ni- 
 trate of bismuth is soluble in a small amount of water, but is 
 decomposed by large amounts, an insoluble basic nitrate be- 
 ing formed. The nitrates of tin are also decomposed by water. 
 
 3. All normal cJilorides except silver, mercurous and 
 cuprous chlorides, and the double platinum-chlorides of 
 potassium and ammonium are soluble. Chloride of lead 
 dissolves but sparingly in cold water. Stannous chloride 
 is soluble in a small amount of water, but decomposes in the 
 presence of large amounts ; the same is true of the chlorides 
 of bismuth and antimony, insoluble basic chlorides being 
 formed if much water is present. 
 
 4. All normal bromides are soluble except silver, mercu- 
 rous and cuprous bromide. Bromide of lead and mercurous 
 bromide are with difficulty soluble. Bismuth and antimony 
 bromide are decomposed by water. 
 
 5. Iodides behave in a similar manner to bromides. 
 Iodide of lead is even less soluble than bromide. 
 
 6. Sulphates. The sulphates of barium, strontium and 
 lead and several basic sulphates are insoluble in water. 
 Sulphates of antimony and bismuth are decomposed, form- 
 ing insoluble basic compounds. Mercurous, silver and cal- 
 cium sulphate are soluble with great difficulty. 
 
 7. All the sulphides of sodium, potassium and ammo- 
 nium are soluble. The sulphides of calcium, magnesium 
 and barium prepared in the dry way are soluble with great 
 difficulty. The polysulphides and sulphydrates of these 
 elements are quite readily soluble. 
 
 8. None of the normal carbonates, arseniates, arsenites. 
 
122 AN AQUEOUS SOLUTION. [§ 88. 
 
 phosphates or borates besides those of the alkali metals are 
 readily soluble in water. Several acid phosphates and acid 
 carbonates are readily soluble. 
 
 9. A few cyanides, oxalates, tartrates, chromates and 
 sulphites, besides those of the alkali metals already men- 
 tioned, are soluble. 
 
 It is obvious that any element of the thirty-six considered 
 in this treatise may be present in an aqueous solution ; but 
 it is also evident from the above list that a great number 
 of salts are absolutely excluded because of their insolubility 
 in water. 
 
 If, on the contrary, there is no certainty that the substance 
 under examination is not a complex artificial mixture, few 
 conclusions can be safely drawn from the fact that a part of 
 it or the whole of it dissolves in water. 
 
 Test the solution with litmus paper. The solution is 
 either neutral, acid or alkaline. 
 
 a. The solution is neutral. The normal salts of most of 
 the metals have an acid reaction. Sodium, potassium, ba- 
 rium, strontium, calcium, magnesium, manganese and sil- 
 ver are the only metallic elements which form salts whose 
 solutions are neutral. Add to two or three drops of the 
 solution a drop or two of carbonate of sodium. If a pre- 
 cipitate forms, any of the above-mentioned elements may be 
 present, and are to be tested for in regular course according 
 to the methods of Part I; if no precipitation ensues, only 
 sodium and potassium or ammonium can be present, and 
 they may be tested for directly, as described in Chap. VIII. 
 
 b. The solution is acid. The acidity may be caused by 
 a normal salt having an acid reaction, or by an acid salt. 
 
 Neither carbonates nor sulphides can be present in an 
 aqueous solution with an acid reaction. The solution is to 
 be tested for the metallic elements in the usual manner. 
 
 c. The solution is alkaline. The alkalinity may be due 
 to the hydrates, sulphides, cyanides or carbonates of the 
 
§ 88.] AN AQUEOUS SOLUTION. 123 
 
 metals belonging to Classes VI and VII ; to the presence of 
 a borate, silicate, phosphate, arseniate, arsenite or alumi- 
 nate of sodium or potassium ; to free ammonia or to carbo- 
 nate of ammonium. If the alkalinity proceed from an 
 alkaline sulphide, the metals whose sulphides are insoluble 
 in water and in alkaline sulphides must be absent. If it 
 is due to the presence of the hydrates or carbonates of the 
 metals of Classes VI and VII, a very large number of sub- 
 stances are excluded. If it proceed from ammonia or car- 
 bonate of ammonium, all substances precipitable by these 
 reagents are absent. 
 
 An alkaline solution may obviously contain some sub- 
 stance, soluble in an alkaline solvent like sodium hydrate 
 or sulphide of ammonium, but liable to immediate precipi- 
 tation when this solvent is destroyed by the addition of 
 hydrochloric acid at the first step of the analysis. Thus 
 any sulphide of Class III dissolved in caustic soda or in an 
 alkaline sulphide, or compounds of alkaline hydrates with 
 the hydrates of aluminum, zinc or chromium, would be pre- 
 cipitated when the alkaline solvd!it was neutralized. 
 
 Again, the alkaline solution of a silicate of sodium or 
 potassium, when neutralized with acid, yields a very gelat- 
 inous whitish preci]3itate of hydrated silicic acid. From a 
 very concentrated solution of a borate, boracic acid separates 
 in colorless, shining, flat crystals, when the solution is 
 acidified with hydrochloric acid ; but the boracic acid thus 
 separated dissolves when the solution is diluted. Again 
 (although, of course, this could not happen in the case of 
 an alkaline solution actually made by dissolving a solid sub- 
 stance in water), chloride of silver dissolved in ammonia- 
 water would be thrown down by any acid added in excess 
 to the solution. (§ 17, page 21.) An alkaline solution 
 containing a poly sulphide or a thiosulphate would give a 
 precipitate of sulphur on acidifying it. 
 
 In view of these possibilities, an alkaline aqueous solution 
 
124 AN AQUEOUS SOLUTION. [§ 88. 
 
 should be carefully neutralized with nitric acid, as a pre- 
 liminary measure, before hydrochloric acid is added to it. 
 Effervescence should be watched for, and, if it occurs, 
 studied as directed in §§ 57-62. Several different cases of 
 precipitation may be distinguished, requiring somewhat dif- 
 ferent treatment. - 
 ^*— ^ a. If the characteristic gelatinous precipitate of silicic 
 I acid appears, the acidulated solution must be evaporated to 
 I dryness and ignited. The silicic acid is thus rendered 
 I insoluble. The ignited residue is digested with dilute nitric 
 I acid and filtered. The filtrate is ready for the usual course 
 1 of analysis. The insoluble residue is silicic acid. 
 
 b. If the glistening colorless plates of boracic acid ap- 
 pear, dilution with warm water will cause them to redissolve. 
 
 c. If a precipitate appear on neutralization, whose color 
 or texture proves that it is neither silicic nor boracic acid, 
 but some substance insoluble in water and dilute acids, 
 thrown down in consequence of the destruction of its alka- 
 line solvent, the liquid is made slightly acid and then fil- 
 tered. The filtrate is ready for the usual course of analysis. 
 
 The precipitate, rinsed with a little water, is reserved 
 for further treatment ; it is not properly a substance soluble 
 in water, and it must be brought into solution by other 
 methods, hereafter to be described. Sometimes a precipi- 
 tate forms on exact neutralization of an alkaline fluid, 
 which redissolves when the acid is added in excess. 
 
 II. An Acid Solution. — Of the three kinds of acid solu- 
 tions described in § 87, any one, any two, or all three, may 
 be obtained from a single mixture of different solids. 
 There is an advantage in knowing that a part of a complex 
 mixture is soluble in water, a part in hydrochloric acid, a 
 part in nitric acid and a part only in aqua regia; because 
 this knowledge may enable the student, when he has found 
 out all the elements of the mixture, to make a more 
 probable guess at the manner of their combination in the 
 
§ 88, 89.] AN ACID SOLUTION. 125 
 
 original mixture, than he would otherwise be able to. But 
 it is usually unnecessary to keep the three kinds of acid 
 solution apart, when all three have been obtained, and to 
 analyze them separately. On the contrary, all three should 
 be mixed together, and analyzed in one course of testing. 
 It must only be borne in mind that when lead, silver, or 
 mercurous salts are present, the nitric acid solution of the 
 residue from the aqueous solution will give a precipitate 
 of the insoluble chlorides of Class I, on being mixed with a 
 hydrochloric acid or aqua regia solution, and also that in 
 some cases the mixing of the several solutions may cause the 
 formation of an insoluble precipitate by reactions between 
 bodies in solution. It is therefore advisable to mix a small 
 portion of the various solutions and see whether any tur- 
 bidity or precipitate is formed. If none appears, the solu- 
 tions may be safely mixed and the analysis proceeded with. 
 If^ however, an insoluble precipitate does appear, the several 
 solutions must be analyzed separately. 
 
 The student must be careful to use no more acid than is 
 absolutely essential. Nitric acid, particularly, is very ob- 
 jectionable ; because when free it reacts upon sulphuretted 
 hydrogen with mutual decomposition, sulphur being set 
 free. It has, therefore, been already prescribed to remove 
 the free acid by evaporation. Sometimes a strongly acid 
 solution becomes turbid when merely diluted with water. 
 This phenomenon points to the presence of bismuth or 
 antimony. The turbidity will disappear again on the addi- 
 tion of hydrochloric acid. 
 
 The mixed acid solutions, after the filtration from any 
 precipitated chlorides of Class I, as first mentioned, are 
 submitted to the regular course of analysis for the metallic 
 elements. 
 
 89. Examination of the Solutions for the Non-metallic 
 Elements. — As has been already stated in § 43, the exami- 
 nation for the non-metallic elements generally follows that 
 
126 TESTS FOR NON-METALLIC ELEMENTS. [§ 89. 
 
 for tlie metallic elements. The method of procedure is 
 usually as follows : — If the substance is soluble in water 
 and there is reason to believe that it is not a complex arti- 
 ficial mixture, but a simple substance, the determination 
 of the metallic element generally proclaims the absence of 
 certain classes of salts, so that it is very seldom necessary 
 to apply more than the barium and the silver tests and a 
 few special tests. This will best be illustrated by taking 
 two examples. 
 
 Suppose a homogeneous solid, which dissolves readily in 
 water, and which proves to contain strontium. Of the 
 classes of salts coming within the range of this manual 
 (see p. 78), only the sulphide, chloride, bromide, iodide, 
 cyanide, nitrate, chlorate and acetate of strontium are sol- 
 uble in water; the presence or absence of a sulphide (and 
 probably of a cyanide) will have been shown when hydro- 
 chloric acid was added to precipitate the members of 
 Class I; the silver test will show either the presence or 
 absence of the chloride, bromide, iodide and cyanide, and 
 these various classes of salts, if any are present, must be 
 distinguished by special tests; special tests must also be 
 applied for nitrates and chlorates, and also for acetates if the 
 substance blackened in the closed tube. In this case, then, 
 the silver test is the only general test necessary, and the num- 
 ber of special tests should hardly be more than four or five. 
 
 Again, suppose the substance soluble in water proves to 
 contain a mercurous salt, the only classes of salts to be 
 sought for would be the sulphate (to be decided by the 
 barium test), the cyanide, chlorate, nitrate and acetate (to 
 be determined by special test). 
 
 If the only metallic element found in an aqueous solution 
 were sodium or potassium (or the radical ammonium), it 
 would be necessary to look for all the classes of salts enu- 
 merated on page 78. In this case the barium test would be 
 applied first, then the silver test, then such special tests as 
 
§ 89.] TESTS FOE NON-METALLIC ELEMENTS. 127 
 
 had not been rendered unnecessary by negative evidence 
 obtained from the general tests. The calcium test would 
 also be applied if the barium test had made the presence of 
 an oxalate, tartrate or fluoride not impossible. 
 
 In the case of an acid solution the conclusions to be 
 drawn from the presence of certain metallic elements are 
 not so general, still in the case of simple substances the ab- 
 sence of certain classes of salts will generally be made sure. 
 For example, a substance under examination, presumptively 
 a simple salt, proves to be insoluble in water, soluble in 
 hydrochloric acid and to contain nickel. The sulphate, chlo- 
 ride, bromide, iodide, chlorate, acetate and nitrate of nickel 
 are soluble in water, and for this reason are excluded from 
 consideration; the sulphite, hyposulphite, sulphide, arseni- 
 ate, arsenite and carbonate would have revealed themselves 
 in the course of the examination for the metallic elements, 
 and the only salts to be specially tested for at this point 
 are the phosphate, oxalate, tartrate and silicate. If the 
 general and special tests fail to indicate the presence of 
 any of the classes of salts mentioned on page 78, the sub- 
 stance may be an oxide or hydrate. The hydrates give off 
 water when heated in a closed tube ; peroxides sometimes 
 give off oxygen under the same conditions, and most ox- 
 ides, when mixed with an excess of carbon and heated, give 
 off carbonic oxide, or carbonic acid gas ; these gases may be 
 recognized as stated on page 95 under oxalates. 
 
 It is evident from what has just been said, that a knowl- 
 edge of the solubility of chemical compounds is of great 
 value in determining the presence or absence of various 
 classes of salts. The student of qualitative analysis should 
 always have at hand a copy of some work on general chem- 
 istry to which he can turn ; but for convenience of reference, 
 a table will be found on pages 128, 129, in which are stated 
 in a general way the solubilities of the more commonly 
 occurring salts. 
 
128 SOLUBILITIES OF CHEMICAL COMPOUNDS. [§ 89. 
 
 Table of 
 
 
 
 Al [NH4] 
 
 |Sb 
 
 As 
 
 Ba 
 
 Bi 
 
 Cd 
 
 Ca 
 
 Cr"i 
 
 Co 
 
 Cu 
 
 Au 
 
 Pe» 
 
 Fe^ 
 
 Acetate 
 
 W, 
 
 W 
 
 u 
 
 U 
 
 W 
 
 W 
 
 W 
 
 W 
 
 Wi 
 
 W 
 
 Wi 
 
 U 
 
 W 
 
 Wi 
 
 Arseniate 
 
 A 
 
 W 
 
 u 
 
 u 
 
 A 
 
 (A) 
 
 U 
 
 A 
 
 A 
 
 A 
 
 A 
 
 u 
 
 A 
 
 A 
 
 Arsenite 
 
 U 
 
 w 
 
 u 
 
 u 
 
 (W) 
 
 U 
 
 u 
 
 A 
 
 U 
 
 A 
 
 A. 
 
 u 
 
 A 
 
 A 
 
 Borate 
 
 u 
 
 w 
 
 u 
 
 u 
 
 (W) 
 A 
 
 A 
 
 (W) 
 
 (W) 
 
 A 
 
 (W) 
 A 
 
 (W) 
 A 
 
 U 
 
 A 
 
 A 
 
 Bromide 
 
 w 
 
 w 
 
 (We) (W) 
 
 W 
 
 (We) 
 
 W 
 
 W 
 
 W 
 
 W 
 
 W4 
 
 W 
 
 W 
 
 W 
 
 
 
 
 Al 
 
 A 
 
 
 Al 
 
 
 
 
 
 
 
 
 
 Carbonate 
 
 u 
 
 w 
 
 u 
 
 u 
 
 A 
 
 A 
 
 A 
 
 A 
 
 A 
 
 A 
 
 A 
 
 u 
 
 A 
 
 A 
 
 Chlorate 
 
 w 
 
 w 
 
 u 
 
 u 
 
 W 
 
 u 
 
 W 
 
 W 
 
 u 
 
 W 
 
 W 
 
 u 
 
 W 
 
 W 
 
 Chloride 
 
 w 
 
 w 
 
 "■1: 
 
 A 
 
 W 
 
 We 
 
 Al 
 
 W 
 
 W 
 
 w 
 
 W 
 
 w, 
 
 w 
 
 W 
 
 W 
 
 Chromate 
 
 A 
 
 w 
 
 A 
 
 U 
 
 A 
 
 A 
 
 (W) 
 A 
 
 (W) 
 
 A 
 
 A 
 
 (W) 
 
 u 
 
 U 
 
 A 
 
 Cyanide 
 
 u 
 
 w 
 
 u 
 
 U 
 
 (W) 
 
 U 
 
 (W) 
 
 w 
 
 A 
 
 A 
 
 A 
 
 W5 
 
 (A) 
 
 W 
 
 Fluoride 
 
 1 
 
 w 
 
 w 
 
 W 
 
 (W) 
 
 W 
 
 (W) 
 
 I 
 
 W 
 
 (W) 
 
 (W) 
 
 u' 
 
 (W) 
 
 (W) 
 
 Hydrate 
 
 A 
 
 w 
 
 A-I 
 
 A 
 
 (W) 
 
 A 
 
 A 
 
 (W) 
 
 A 
 
 A 
 
 A 
 
 A 
 
 A 
 
 A 
 
 ''lu:'- 
 
 U 
 
 w 
 
 u 
 
 u 
 
 (W) 
 A 
 
 U 
 
 W 
 
 w 
 
 U 
 
 W 
 
 U 
 
 U 
 
 W 
 
 U 
 
 Iodide 
 
 w 
 
 w 
 
 (We) (W) 
 
 w 
 
 A 
 
 W 
 
 w 
 
 W 
 
 W 
 
 U4 
 
 A 
 
 W 
 
 w 
 
 Nitrate 
 
 w 
 
 w 
 
 u 
 
 U 
 
 w 
 
 We 
 A, 
 
 W 
 
 w 
 
 w 
 
 W 
 
 w 
 
 W 
 
 W 
 
 w 
 
 Oxalate 
 
 A 
 
 w 
 
 u 
 
 u 
 
 A 
 
 -"1 
 A 
 
 A 
 
 A 
 
 w 
 
 A 
 
 A 
 
 u 
 
 A 
 
 A 
 
 Oxide 
 
 A 
 
 w 
 
 A 
 
 w 
 
 W 
 
 A 
 
 A 
 
 (W) 
 
 A-I 
 
 A 
 
 A 
 
 A 
 
 A 
 
 A 
 
 Phosphate 
 
 ! A 
 
 w 
 
 (W) 
 A 
 
 u 
 
 Aj 
 
 A 
 
 A 
 
 A2 
 
 A 
 
 A 
 
 A 
 
 u 
 
 A 
 
 A 
 
 Silicate 
 
 A-I 
 
 u 
 
 U 
 
 u 
 
 A-I 
 
 U 
 
 U 
 
 A-I 
 
 U 
 
 U 
 
 A-I 
 
 U 
 
 (A)-I (A).I 
 
 Sulphate 
 
 W 
 
 w 
 
 A 
 
 u 
 
 I 
 
 A 
 
 W 
 
 (W) 
 A 
 
 w 
 
 w 
 
 W 
 
 u 
 
 W 
 
 Wi 
 
 Sulphide 
 
 u 
 
 w 
 
 A 
 
 A 
 
 W 
 
 A 
 
 A 
 
 (W)8 
 
 u 
 
 A 
 
 A 
 
 A 
 
 A 
 
 u 
 
 Sulphite 
 
 W 
 
 w 
 
 U 
 
 U 
 
 A 
 
 A 
 
 A 
 
 (W) 
 A 
 
 A 
 
 A 
 
 A 
 
 U 
 
 (W) 
 
 A 
 
 Tartrate 
 
 W 
 
 w 
 
 W 
 
 u 
 
 (W) 
 A 
 
 A 
 
 (W) 
 
 (W) 
 
 W 
 
 W 
 
 (W) 
 
 U 
 
 (W) 
 
 W 
 
 
 Al [NH4; 
 
 |Sb 
 
 As 
 
 Ba 
 
 Bi 
 
 Cd 
 
 Ca 
 
 Cr»i 
 
 Co 
 
 Cu 
 
 Au 
 
 Fe« 
 
 Fe"» 
 
 ' 
 
 In this table W signifies that the substance is readiljr soluble in water, 
 (W) that the substance is soluble with difficulty in water, ^ . ^ that the 
 substance is sparingly soluble in water, readily soluble in acids ; A and 
 (A) that the substance is readily or with difficulty dissolved by acids ; I 
 that the substance is insoluble in water or acids ; U signifies that the 
 compound is either unknown or so uncommon as rarely to be mej^with. 
 
§89.] SOLUBILITIES OF CHEMICAL COMPOUNDS. 129 
 Solubilities. 
 
 
 Pb Mg 
 
 Mn 
 
 Hgr> 
 
 Hg" 
 
 Ni 
 
 Pt 
 
 K 
 
 Agr 
 
 Na 
 
 Sr 
 
 Sii« 
 
 Sn'» 
 
 Zn 
 
 
 W 
 A 
 A 
 A 
 
 w 
 
 A 
 A 
 A 
 
 W 
 A 
 A 
 A 
 
 (W) 
 A 
 A 
 U 
 
 W 
 A 
 A 
 U 
 
 W 
 A 
 A 
 A 
 
 u 
 u 
 u 
 u 
 
 W 
 W 
 W 
 
 w 
 
 (W) 
 
 A 
 
 A 
 
 (W) 
 A 
 
 W 
 W 
 
 w 
 
 w 
 
 A 
 
 A 
 
 (W) 
 
 w 
 
 A 
 A 
 A 
 
 W 
 
 A 
 A 
 
 U 
 
 W 
 A 
 
 U 
 A 
 
 Acetate 
 
 Arseuiate 
 
 Arsenite 
 
 Borate 
 
 (W) 
 
 W 
 
 W 
 
 A 
 
 (W) 
 
 W 
 
 w 
 
 w 
 
 (A) 
 
 w 
 
 w 
 
 W 
 
 w 
 
 W 
 
 Bromide 
 
 A 
 W 
 
 (W) 
 
 A 
 W 
 W 
 
 A 
 
 U 
 
 w 
 
 A 
 W 
 
 (A) 
 
 A 
 W 
 W 
 
 A 
 W 
 W 
 
 A 
 
 u 
 w 
 
 w 
 w 
 
 A 
 W 
 
 (A)-I 
 
 w 
 w 
 
 A 
 W 
 W 
 
 A 
 W 
 W 
 
 A 
 W 
 W 
 
 A 
 \V 
 W 
 
 Carbonate 
 
 Chlorate 
 
 Chloride 
 
 A 
 
 W 
 
 u 
 
 A 
 
 (W) 
 
 A 
 
 u 
 
 w^ 
 
 A 
 
 w 
 
 (W) 
 
 A 
 
 u 
 
 W, 
 
 Chromate 
 
 A 
 A 
 
 (W) 
 
 W 
 
 (AJ 
 
 A 
 
 A 
 A 
 
 A 
 
 U 
 A 
 
 A 
 
 w 
 w 
 
 A 
 
 (A) 
 (W) 
 
 A 
 
 A 
 
 (A) 
 
 w 
 
 A 
 
 w 
 w 
 
 w 
 
 A 
 W 
 
 A 
 
 w 
 w 
 
 w 
 
 u 
 
 (W) 
 (W) 
 
 u 
 w 
 
 A 
 
 u 
 w 
 
 A 
 
 A Cyanide 
 (W) Fluoride 
 A 
 A Hydrate 
 
 (W) 
 A 
 
 (W) 
 
 W 
 W 
 
 W 
 W 
 
 U 
 A 
 
 U 
 A 
 
 W 
 W 
 
 U 
 I 
 
 w 
 w 
 
 (W) 
 A 
 A 
 
 w 
 w 
 
 w 
 w 
 
 W 
 
 (W) 
 
 u 
 
 (W) 
 
 U 
 W 
 
 Iodide 
 
 w 
 
 W 
 
 W 
 
 W, 
 
 w, 
 
 W 
 
 W 
 
 w 
 
 W 
 
 w 
 
 w 
 
 A 
 
 A 
 
 W 
 
 Nitrate 
 
 A 
 
 (W) (W) 
 
 A 
 
 A 
 
 A 
 
 W 
 
 w. 
 
 (W) 
 
 w 
 
 A 
 
 A 
 
 W 
 
 A 
 
 Oxalate 
 
 A 
 A 
 
 A 
 A 
 
 A 
 A 
 
 A 
 A 
 
 A 
 A 
 
 A 
 A 
 
 A 
 U 
 
 w 
 w 
 
 A 
 A 
 
 w 
 w 
 
 (W) 
 A, 
 
 A 
 
 A 
 
 I 
 A 
 
 A 
 A 
 
 Oxide 
 Phosphate 
 
 U 
 
 I 
 
 I 
 W 
 
 (A)-] 
 W 
 
 : u 
 
 (W) 
 
 U 
 W 
 
 U 
 W 
 
 U 
 W 
 
 w 
 
 U 
 
 (W) 
 
 w 
 w 
 
 A 
 
 I 
 
 U 
 W 
 
 U 
 W 
 
 A-I Silicate 
 W Sulphate 
 
 A 
 A 
 
 (W) 
 A 
 
 A 
 
 (W) 
 A 
 
 A 
 W 
 
 A 
 W 
 
 A 
 A 
 
 A 
 
 u 
 
 w 
 w 
 
 A 
 A 
 
 w 
 w 
 
 W 
 A 
 
 A 
 W 
 
 A 
 
 u 
 
 A 
 A 
 
 Sulphide 
 Sulphite 
 
 A 
 
 W 
 
 (W) 
 
 A 
 
 (W) 
 
 A 
 
 U 
 
 W3 
 
 A 
 
 w 
 
 W 
 
 (W) 
 
 A 
 
 (W) Tartrate 
 A 
 
 Pb Mgr 
 
 Mn 
 
 Hgi 
 
 Hr* 
 
 Ni 
 
 Pt 
 
 K 
 
 Agr 
 
 Na 
 
 Sr 
 
 Sn" 
 
 Sn'- 
 
 Zn 
 
 
 
 1. The basic salt is A. 6. The salt is soluble in water, but decomposed 
 
 2. The acid salt is W, by large amounts, insoluble basic com- 
 
 3. The acid salt is (W). pounds being formed. 
 
 4. The -ous salt is A. 7. Is decomposed by boiling with water. 
 
 5. The -ous salt is I. 8. The poly sulphides and sulphydrates are (W). 
 
130 INSOLUBLE SUBSTANCES. [§§ 89, 90. 
 
 It is to be distinctly borne in mind that this table is not 
 exhaustive, and is intended merely as a help to the begin- 
 ner, who is analyzing comparatively simple substances. 
 In the case of complex mixtures the table is of little ser- 
 vice. Thus, chloride of silver is designated as I, but it 
 would be easy to mix chloride of silver and chloride of 
 sodium in such proportions that on treatment with water 
 the whole of the mixture would go into solution : so too the 
 chlorides of potassium and platinum are readily soluble in 
 water, each by itself, but the double chloride of these two 
 elements is so insoluble that it is used as a test for potas- 
 sium (§ 43). 
 
 D. Treatment of Insoluble Substances. 
 
 90. The substances of common occurrence which are prac- 
 tically insoluble in water and acids are : — 
 
 The sulphates of barium, strontium and lead. Sometimes 
 sulphate of calcium occurs in difficultly soluble forms. 
 
 Chloride of silver. 
 
 The anhydrous sesquioxides of aluminum, chromium and 
 iron, either native, or the result of intense ignition. 
 
 Chrome-iron-ore, a native mineral. 
 
 Some aluminates. 
 
 Binoxide of tin, native, or the result of ignition. 
 
 Silica and many silicates. 
 
 Fluoride of calcium (fluor-spar). 
 
 Beside the substances included in this list, sulphur and 
 carbon, or graphite, should, perhaps, be mentioned, because 
 they are insoluble. Bromide, iodide and cyanide of silver 
 are decomposed by boiling with aqua regia, and converted 
 into the chloride, so that these substances never appear in 
 their proper form in the final insoluble residue. But only 
 in case aqua regia having failed to dissolve the substance, it 
 is directly examined. In case of dealing with a complex 
 
§§90,91.] INSOLUBLE SUBSTANCES. 131 
 
 substance, great care should be taken that all soluble material 
 is removed, otherwise many complications may be introduced. 
 
 91. Substances which resist solution in liquids are gen- 
 erally liquefied by the action of fluxes at a high temperature ; 
 they are fused in contact with some powerful decomposing 
 agent, like the carbonate or acid sulphate of an alkali metal, 
 or the hydrate or carbonate of an alkaline -earth metal. Cer- 
 tain preliminary experiments should precede the fusion. 
 
 The insoluble powder is first examined carefully (with the 
 help of a lens, if convenient) to ascertain if it is a homo- 
 geneous substance of the same color throughout, or a mixture 
 composed of dissimilar, variously colored particles. The 
 following blowpipe experiments sometimes give decisive 
 indications, particularly with homogeneous substances. 
 
 a. The closed-tube test is repeated, looking especially for 
 sulphur. 
 
 b. If the substance is black, and not changed by heating 
 in the closed tube, the presence of carbon in some form is 
 indicated. Heat a small sample on platinum foil ; if it is 
 consumed, carbon is present. Graphite is not consumed, 
 but can be recognized by its property of soiling the fingers 
 or paper. 
 
 c. The reduction test (§ 83) is repeated with great care, 
 looking especially for silver, lead and tin, and applying to 
 the globule, if any is obtained, the test for distinguishing 
 between these three white metals. This test has already 
 been applied to the original substance ; but if this substance 
 was a complex mixture containing soluble ingredients, it 
 is quite possible that the test should give a more satisfac- 
 tory result, now that all substances soluble in water and 
 acids have been removed, than it yielded before. If, how- 
 ever, decided indications of the presence of a reducible 
 metal were obtained in the first instance, the repetition of 
 the test is, of course, unnecessary. If any reducible metal 
 is detected, it is necessary to use a porcelain crucible for the 
 
182 INSOLUBLE SUBSTANCES. [§ 91. 
 
 fusion which it may be desirable to make (§ 92) in order 
 to convert the insoluble substance into a more manageable 
 form. A platinum crucible or piece of platinum foil, which 
 is employed for most fusions, cannot be Used with safety 
 when the substance to be fused contains any reducible 
 mgtal ; for many of the alloys of platinum are extremely 
 fusible. 
 
 d. Prepare another pellet of a mixture of equal parts 
 of the insoluble powder and carbonate of sodium, adding 
 a little charocal powder to the paste. Fuse this mixture 
 upon charcoal in the reducing flame of the blowpipe. Scoop 
 out the fused mass and the surrounding charcoal with a 
 penknife, place the dry mass -upon a bright surface of silver 
 (coin or foil), and wet it with a drop of water. If a brown 
 stain be produced on the silver, it is evidence of the pres- 
 ence of sulphide of sodium in the fused mass. This sul- 
 phide results from the reduction of a sulphate, and is 
 evidence of the presence of a sulphate in the substance 
 tested. The odor of sulphuretted hydrogen is often per- 
 ceptible when the fused mass is moistened. The silver 
 coin or foil may be replaced by a piece of lead paper, if care 
 be taken not to mistake the mere dirtying of the paper for 
 a stain of sulphide; the silver is, however, to be preferred; 
 it may be cleaned after use by treating it with a solution 
 of cyanide of potassium and then washing with water. It 
 is obvious that the carbonate of sodium used in this test 
 must be so free from sulphate of sodium as not itself to 
 give this reaction on silver, after fusion on charcoal. Since 
 coal gas invariably contains traces of sulphur compounds, 
 the test cannot be performed with a gas flame ; a candle or 
 lamp flame (App., § 82) must be employed. 
 
 It is, of course, possible to apply this test for sulphates 
 to the fused mass obtained in (c) ; this second test is in fact 
 unnecessary when a decided reaction has been obtained in 
 the first instance. 
 
§ 91.] INSOLUBLE SUBSTANCES. 133 
 
 e. Make the loop on the end of the bit of platinum wire 
 (App., § 83) white-hot in the blowpipe flame, and thrust it 
 white-hot into some powdered borax ; a quantity of borax 
 will adhere to the hot wire ; reheat the loop in the oxidiz- 
 ing flame ; the borax will puff up at first, and then fuse to a 
 transparent glass. If enough borax to form a solid, trans- 
 parent bead within the loop does not adhere to the hot wire 
 the first time, the hot loop may be dipped a second time 
 into the powdered borax. 
 
 When a transparent glass has been formed within the 
 loop of the platinum wire, touch the bead of glass while it 
 is hot and soft, to a few particles of the insoluble powder, 
 and reheat the bead with the adhering powder in the oxi- 
 dizing flame; If the substance dissolves slowly in the 
 borax, and the bead has a fine yellowish-green color when 
 cold, chromium is probably present. Reheat the bead in 
 the reducing flame : if it presents a bright green color both 
 when hot and cold, there is no doubt of the presence of 
 chromium. 
 
 It sometimes happens when too much of the substance 
 to be tested has been added, that the borax bead becomes 
 so dark colored as to be practically opaque. It may then 
 be flattened while soft, by sudden pressure between any 
 smooth metallic surfaces, like the flat parts of jewellers' 
 tweezers. If the flattening makes the color of the borax 
 glass visible, -nothing more is necessary; but if the glass 
 is still too dark, all the glass outside the loop of platinum 
 may be broken oif by gentle hammering, and the remaining 
 glass may be reheated and largely diluted by the addition 
 of more borax. 
 
 It is convenient to be informed of the presence of chro- 
 mium, because chromic oxide and chrome iron ore are sub- 
 stances which it is particularly difficult to decompose 
 effectually by fusion. In presence of chromium, no other 
 bead reaction which can be anticipated under the circum- 
 stances will give a decisive result; but in the absence of 
 
134 INSOLUBLE SUBSTANCES. — FUSIONS. [§§ 91, 92. 
 
 chromium, the presence of iron may be determined. A 
 suitable quantity of oxide of iron causes the borax bead, 
 heated in the oxidizing flame, to look red when hot and 
 yellow when cold. In the reducing flame the iron bead 
 becomes greenish, or light brownish-green. 
 
 This test is rendered unnecessary if the substance under 
 examination be a white powder, or if from other appear- 
 ances the absence of chromium is assured. 
 
 /. The test for fluorine ( § 71) should be applied to the 
 original substance, if it has not already been done. 
 
 When all the above-mentioned tests (a-f) give negative 
 results, the simplification of the problem is very conspicuous ; 
 the substances which may be present are reduced to alumina 
 and some aluminates, silica and silicates. Again, in the 
 case of a substance evidently homogeneous, if the prelimi- 
 nary tests give aflirmative results, the indications of the 
 character of the substance are almost conclusive. Thus 
 chloride of silver, sulphate of lead, chromic or ferric oxide, 
 binoxide of tin, or fluoride of calcium, may be satisfactorily 
 identified by the preliminary tests alone. 
 
 There are two methods of changing insoluble substances 
 into more manageable forms by the application of heat with 
 sufficient exactness for the purposes of the qualitative 
 analyst, — the method of fusion, and the method by defla- 
 gration. 
 
 92. Fusions. — Mix the fine powder of the insoluble sub- 
 stance with about six parts by weight of dry carbonate of 
 sodium in powder. If free sulphur is present, it must be 
 first removed by gentle ignition on a piece of porcelain. 
 Both powders must be as fine as they can be made, and 
 they must be intimately mixed. Keep the mixture at a 
 bright red heat, on a piece of platinum foil or in a platinum 
 crucible ( a porcelain crucible, if a reducible metal has been 
 found in the substance, § 91, c, and fusion is for any reason 
 preferred to deflagration ), until the mass has been brought 
 
§§92,93.] PUSlON OF INSOLUBLE SUBSTANCES. 135 
 
 to a state of quiet fusion (App., § 79). Place the hot plat- 
 inum crucible, when withdrawn from the lamp or fire, on a 
 cold block, or thick plate of iron, and let it cool. If a gas 
 blast-lamp be employed, the supply of gas may be inter- 
 rupted, and the blast of air, directed as before, upbn the 
 crucible, until it has become cold. 
 
 When the green borax bead, and the dark color of the 
 insoluble powder, point to the presence of chrome iron ore, 
 a mixture of three parts, by weight, of carbonate of sodium 
 with three parts, by weight, of nitre, may be substituted 
 for the six parts of carbonate of sodium alone. 
 
 93. Treatment of the Fused Mass. — When the crucible 
 has been cooled in the way mentioned above, the fused mass 
 can generally be removed from the crucible in an unbroken 
 lump. Soak the lump in boiling water until everything is 
 dissolved which is soluble in water. If the mass cannot be 
 detached from the crucible, the crucible and its contents 
 must be soaked in boiling water. 
 
 The aqueous solution of the fused mass is filtered from 
 the residue insoluble in water and reserved. That portion 
 of the fused mass which boiling water did not dissolve is 
 treated with acid, — hydrochloric acid if silver and lead be 
 absent, nitric acid if either of these metals be present, — 
 and the acid solution obtained is treated as will be de- 
 scribed. If a portion of the fused mass resist both water 
 and acids, the insoluble portion may consist of separated 
 silicic acid, or of some of the original substance undecom- 
 posed by the fusion. In the latter case, another and more 
 prolonged fusion is the oiily effectual remedy, although it 
 may often happen that a partial decomposition of the 
 insoluble substance will enable the analyst to recognize all 
 the elements which it contains. 
 
 The treatment of the aqueous and acid solutions of the 
 last paragraph will be best understood if we first consider 
 what changes take place in the process of fusion. Suppose 
 
136 FUSION OF INSOLUBLE SUBSTANCES, [§ 93. 
 
 that the substance in hand is sulphate of barium ; at the high 
 temperature of the fusion the barium and sodium change 
 places and, instead of sulphate of barium and carbonate 
 of sodium, there result carbonate of barium and sulphate 
 of sodium : BaSO^ + Na^COg = NagSO^ + BaCOg. When the 
 fused mass is treated with water, the sulphate of sodium 
 (with the excess of carbonate of sodium) goes into solution, 
 while the carbonate of barium, which is insoluble in water, 
 is dissolved by the hydrochloric (or nitric) acid as chloride 
 (or nitrate) of barium. Therefore in a case like this, the 
 metallic element would be found in the acid solution and 
 the class or kind of salt might be determined by applying 
 the test for sulphates to the aqueous solution. Again, sup- 
 pose the substance under examination is a double silicate 
 of calcium and aluminum. After fusion with carbonate of 
 sodium and treatment successively with water and acid, a 
 portion of the silicic acid will be in the aqueous solution 
 (as silicate of sodium) and a portion in the acid solution, 
 while a part may remain insoluble ; the aluminum will ex- 
 ist partly in the aqueous solution (as aluminate of sodium) 
 and partly in the acid solution (as chloride of aluminum) ; 
 the calcium, which after the fusion remained as car- 
 bonate, insoluble in water, will have been converted into 
 chloride and will be found in the acid solution. Having 
 considered the general nature of the changes brought about 
 by fusion with carbonate of sodium, we proceed to the state- 
 ment of the treatment of the aqueous and acid solutions of 
 the fused mass. 
 
 a. If the test for sulphates described in § 91, d, on page 
 132, has failed to give satisfactory indications, acidify a 
 small portion of the aqueous solution with hydrochloric 
 acid, and apply the barium test. The carbonate of sodium 
 used in the fusion should be free from sulphate. 
 
 6. Acidify another small portion with acetic acid, and 
 apply the lead test for chromates (§ 63). In presence of 
 
§ 93.] FUSION OF INSOLUBLE SUBSTANCES. 137 
 
 sulphuric acid this test will be obscured, but not rendered 
 wholly useless (§ 31, p. 49). 
 
 c. Acidify a third portion with nitric acid, and apply 
 the silver test for chlorine. The student must first prove 
 that his carbonate of sodium contains no chloride. 
 
 d. If the test for fluorine by the method of § 71 has for 
 any reason been unsatisfactory, a fourth portion, having 
 been concentrated by evaporation in a porcelain dish, and 
 again cooled, is acidified with hydrochloric acid, and then 
 left at rest until the carbonic acid has escaped. It is then 
 supersaturated with ammonia, heated, and filtered while 
 hot. The filtrate is collected in a bottle; chloride of cal- 
 cium is immediately added to it; the bottle is closed and 
 allowed to stand at rest. If the original substance con- 
 tained a fluoride, the fluorine will have combined with 
 sodium during the fusion, and fluoride of sodium will be 
 contained in the aqueous solution. The carbonic acid hav- 
 ing been expelled, and all substances precipitable by ammo- 
 nia having been removed, the chloride of calcium will throw 
 down the fluoride of calcium. If a precipitate separates from 
 the liquid in the bottle after some time, it is collected in a 
 small filter, dried and examined for fluorine by the method 
 of § 71. 
 
 When the tests a-d give negative results, or when by pre- 
 vious tests the absence of sulphates, chromates, chlorides 
 and fluorides is made certain, the remainder of the aqueous 
 solution is added to the acid solution and the mixture is 
 evaporated to dryness and ignited; the residue thus ob- 
 tained is boiled with dilute hydrochloric (or nitric) acid. 
 If the dilute acid fails to dissolve the residue completely, 
 tlie insoluble portion consists of silicic acid. The solution is 
 examined in the usual way for the metallic elements (§ 45), 
 except, of course, sodium (and sometimes potassium), which 
 has been added in the flux. (See § 94, p. 139.) 
 
 When the preliminary examination or the tests a-d show 
 
138 FUSION OF INSOLUBLE SUBSTANCES. [§93. 
 
 the presence of one or more of the classes of salts men- 
 tioned above, the treatment is slightly different. In that 
 case the remainder of the aqueous solution is acidified with 
 hydrochloric acid, evaporated to dryness and ignited; the 
 residue thus obtained is boiled with dilute hydrochloric 
 acid. If the acid fails to dissolve the residue completely, 
 the insoluble portion consists of silica; the solution may 
 be tested for aluminum. (See p. 136.) If silicic acid has 
 been found in the aqueous solution, the original acid solu- 
 tion of the fused mass is evaporated to dryness and ignited ; 
 the residue is treated with dilute acid, and the solution, 
 after being filtered from the silicic acid, is examined for the 
 metallic elements in the usual manner. It is evidently im- 
 practicable in this last case to mix the aqueous and the acid 
 solutions, for if the case of sulphate of barium (p. 136) be 
 taken as an example, in mixing the aqueous solution (con- 
 taining sulphate of sodium) and the acid solution (contain- 
 ing chloride of barium), there would be an immediate 
 precipitation of the insoluble sulphate of barium, which 
 was the substance to be analyzed. 
 
 Sil icates are by far the most common of insoluble sub- 
 stances. A great variety of metallic elements occur in 
 insoluble silicious minerals, so that the possible contents of 
 the acid solution of the fused mass are very various. The 
 evaporation to dryness in order to render the silica insoluble 
 is prescribed because the subsequent examination goes on 
 the better for this preliminary removal of silica, which, if 
 left in solution, might create confusion by appearing as a 
 precipitate at almost any stage of the analysis. 
 
 Many silicates contain sodium and potassium. When the 
 presence or absence of these alkali metals is to be deter- 
 mined, it is evident that the pulverized silicate must not be 
 fused with carbonate of sodium, but with some decomposing 
 flux free from alkali. 
 
§§94-96.] INSOLUBLE SUBSTANCES. 139 
 
 94. Decomposition by Means of Carbonate of Calcium and 
 Chloride of Ammonium. — An intimate mixture is prepared 
 of one part of the silicate, six parts of pure precipitated 
 carbonate of calcium, three fourths part of pulverized chlo- 
 ride of ammonium. This mixture is heated to bright redness 
 in a platinum crucible for thirty or forty minutes. The 
 crucible with its contents (which should be in a coherent, 
 sintered, but not thoroughly fused condition) is then placed 
 in a beaker, and soaked for half an hour in water kept near 
 the boiling point. The contents of the beaker are then 
 filtered. The filtrate, containing caustic lime, chloride of 
 calcium and all the sodium and potassium of the original 
 silicate as chlorides, is treated with a little ammonia-water, 
 and with carbonate of ammonium in slight excess ; the liq- 
 uid is heated to boiling and filtered. This second filtrate 
 is evaporated to dryness, and gently ignited to expel the 
 ammonium salts. The residue is dissolved in a little water; 
 one or two drops of carbonate of ammonium and a drop of 
 oxalate of ammonium are then added ; the mixture is again 
 heated and filtered ; this third filtrate is evaporated to dry- 
 ness and ignited; the ignited residue, if there be any, con- 
 sists of the chlorides of sodium and potassium, or of one of 
 these two salts. This residue is examined according to § 43. 
 
 95. Fusion with Acid Sulphate of Sodium. — The fol- 
 lowing method may be tried to advantage upon ferric oxide, 
 chromic oxide or chrome iron ore, and some very refractory 
 silicates. Heat the insoluble substance with three or four 
 times its bulk of acid sulphate of sodium (App., § 30) in a 
 platinum crucible, until the sulphate melts ; then maintain 
 it in the liquid state for half an hour. This operation 
 should be performed under a hood. The fused mass is 
 treated essentially as before (§ 93), allowance being made 
 for the different nature of the flux. 
 
 96. Deflagration. — The method of fusion just described 
 involves the use of a platinum or porcelain crucible, and 
 
140 DEFLA GRA TION. 
 
 demands the heat of a blast-lamp, or strong coal fire. Nei- 
 ther crucibles, lamps nor fires are necessary in the method 
 of deflagration which applies the heat inside the mass to 
 be fused. This decomposition by deflagration is performed 
 as follows : — One part, by weight, of the insoluble powder 
 is intimately mixed with two parts of dry carbonate of so- 
 dium, two parts of fine and pure charcoal powder and twelve 
 parts of powdered nitre. The mixture is put in a thin por- 
 celain dish or clean iron tray; the dish, or little tray, is 
 placed under a hood, or in the open air, and a lighted match 
 is applied to the centre of the heap. The deflagration is 
 completed in two or three seconds, and a well-fused mass 
 remains. This mass is detached from the cooled dish or 
 traj', and boiled with water in a beaker ; it is generally very 
 porous, and is therefore readily disintegrated by stirring it 
 in the hot water with a glass rod. The soluble portion will 
 all be extracted in a very few minutes. The residue left by 
 water is treated with acid precisely as described in § 93. 
 The aqueous and acid solutions of the deflagrated sub- 
 stance are submitted to the same operations as the corre- 
 sponding solutions of substances fused in crucibles. A 
 little charcoal is generally left undissolved by the acid, 
 and with it any of the substance which may have escaped 
 decomposition. The mixture of one part, by weight, of 
 powdered charcoal, and six parts of nitre, may be kept ready 
 mixed for effecting the fusion of insoluble substances. 
 
 The advantages of this process are that it is quick, re- 
 quires only cheap and common tools, and may be applied to 
 substances containing reducible metals, as well as to any 
 others. It is, of course, inapplicable when sodium and 
 potassium are to be sought for in silicates. Chrome iron 
 ore cannot be decomposed in this way. The insoluble 
 sulphates, chloride of silver, binoxide of tin, fluor-spar, 
 glass, and many natural silicates, may be very well treated 
 by this method, in spite of its apparent roughness. 
 
CHAPTER XII. 
 
 TREATMENT OF A PURE METAL OR ALLOY. 
 
 97. The elements which are now used in the arts in the 
 metallic state, and which therefore may come into the hands 
 of the analyst as metals, either pure or alloyed, are silver, 
 lead, mercury, bismuth, cadmium, copper, arsenic, anti- 
 mony, tin, gold, platinum, aluminum, iron, zinc, nickel, 
 manganese and magnesium. These metals can all be brought 
 into solution and detected in the wet way with ease and 
 certainty. It is therefore not worth while to submit a 
 metal, or metallic alloy, to preliminary blowpipe tests, al- 
 though at need mercury and arsenic can be readily detected 
 by the closed-tube test (§ 82), and many others, by exposing 
 them on charcoal to the reducing and oxidizing flame (com- 
 pare § 83, pp. 112-115). 
 
 A portion of the metal or alloy to be examined should 
 first be reduced to as fine a state of division as possible. 
 If it is brittle, it can be powdered; if soft, shavings can 
 be cut from it; if tough and hard, it can perhaps be fused, 
 and shaken into powder while melted, or granulated by 
 being poured from a height into cold water. Filings should 
 be the last resort, because of the possibility of foreign 
 admixture of iron. 
 
 98. Action of Nitric Acid on the Metals. — A small quan- 
 tity of the divided metal or alloy, about the equivalent of 
 a pea in bulk, is placed in a flask, covered with concentrated 
 nitric acid, and heated gently under a hood or in the open 
 air for half an hour or until no further change is noticed 
 on the addition of more concentrated nitric acid. 
 
 141 
 
142 TREATMENT OF ALLOYS. [§ 98. 
 
 If complete solution ensues, gold, platinum, tin and anti- 
 mony are probably altogether absent; they can only be 
 present in very minute proportion. Any of the other metals 
 above enumerated may be present. Transfer the acid solu- 
 tion to a porcelain dish, and evaporate it almost to dryness 
 to drive off the free acid; dilute the evaporated liquid with 
 about ten times its bulk of water, and proceed with the 
 analysis in the usual way (§ 45). If the solution, from 
 which the greater part of the free acid has been removed 
 by evaporation, becomes turbid on the addition of water, 
 bismuth is doubtless present. In this case enough acid must 
 be restored to the solution to clarify it. Mercury, if pres- 
 ent, will be dissolved to mercuric nitrate. 
 
 If a residue remains undissolved, add a little more acid to 
 make sure that the acid is incapable of further action; and 
 when this point is settled, test a drop or two of the clear 
 liquid on platinum foil, to ascertain if anything has entered 
 into solution. If the nitric acid has effected a partial solu- 
 tion of the original metal, evaporate the liquid nearly to 
 dryness, dilute the evaporated mixture with water, filter, 
 and submit the filtrate to the usual course of analysis. The 
 residue is thoroughly washed with water, to prepare it for 
 further treatment. On diluting the evaporated mixture with 
 water, a turbidity due to the presence of bismuth may 
 appear. The experienced operator will hardly fail to dis- 
 tinguish between any such turbidity and a residue insoluble 
 in the nitric acid. To avoid mistakes which would lead to 
 the unnecessary addition of acid, it is well to take out a drop 
 of the nitric acid solution before evaporation and to add 
 it to several teaspoonfuls of water contained in a test- 
 tube. (See p. 26.) The absence or presence of bismuth will 
 thus be discovered. It will sometimes happen that a white 
 residue appears in the nitric acid solution which disappears 
 when the evaporated mixture is diluted. This is owing to 
 the presence of lead: nitrate of lead is rather insoluble in 
 
§ 98.] TREATMENT OF ALLOYS. 143 
 
 stong nitric acid, but passes into solution readily when the 
 mixture is diluted. 
 
 Three different cases of insoluble residues may occur, 
 readily distinguished by the mere appearance of the residue. 
 
 a. The insoluble substance is non-metallic and white. 
 In this case tin and antimony may be present, but gold and 
 platinum are probably absent. The white residue may con- 
 tain the insoluble oxides of tin and antimony, or either of 
 them. These elements are to be detected by the methods 
 of § 25. 
 
 b. The insoluble substance is metallic, as evidenced by 
 its lustre, if it is in visible fragments, or by the weight and 
 gray or black color of its powder, if it is in a fine state of 
 division. Such a residue must be either gold or platinum 
 (or some of the rare platinum-like metals which lie without 
 the range of this manual). The residue is dissolved in 
 aqua regia, and evaporated to a very small bulk. 
 
 Test for Gold. — A portion of this evaporated liquid is 
 diluted with ten times its bulk of water, and poured into a 
 beaker which is placed on a sheet of white paper. A small 
 quantity of a solution of stannous chloride is tinged yellow 
 by the addition of a few drops of solution of ferric chloride 
 (App., § 49), and is then considerably diluted. A glass rod 
 is dipped, first into this tin solution, and then into the so- 
 lution to be tested for gold. If even a trace of the xest 
 precious metal be present, a blue or purple streak for 
 will be observed in the track of the rod. If the •^^• 
 quantity of gold be more considerable, a pink tinge will be 
 imparted to the solution, or a purplish precipitate will be 
 produced by a sufficient quantity of the tin solution. This 
 ^' purple-of-Cassius " test is applicable to very acid solutions. 
 
 Test for Platinum. — To another undiluted por- Test 
 tion of the cooled aqua regia solution, a cold for 
 concentrated solution of chloride of ammonium P*- 
 is added. The formation of a yellow, crystalline precipi- 
 
144 TBEATMENT OF ALLOYS. 
 
 tate of chloroplatinate of ammonium indicates the presence 
 of platinum (or of some rare platinum-like metal). By 
 adding a little alcohol to the liquid, the test is made more 
 delicate. In a difficult case, the aqua regia solution might 
 be evaporated to dryness with chloride of ammonium, and 
 the residue treated with weak alcohol and water, which 
 would dissolve all the ingredients except the chloroplat- 
 inate. Upon ignition, chloroplatinate of ammonium leaves 
 spongy platinum behind. 
 
 It happens exceptionally in the case of certain alloys, 
 especially in the presence of copper, that concentrated^ nitric 
 acid fails to attack them even when it is hot; it is well, 
 therefore, before inferring the presence of an insoluble 
 metallic residue, to try the effect of boiling the seemingly 
 insoluble substance in nitric acid diluted with an equal bulk 
 of water. 
 
 c. The insoluble residue contains both a white powder 
 and a metallic substance. It must then be examined for 
 antimony, tin, gold and platinum precisely as described in 
 §25. 
 
CHAPTER XIII. 
 
 TREATMENT OP LIQUIDS. 
 
 99. Evaporation Test. — The first step in the examina- 
 tion of an unknown liquid is to evaporate a few drops at 
 a gentle heat on platinum foil. Attention should be paid 
 to the smell of the escaping vapors in order to ascertain if 
 the solvent be water or some other fluid, like alcohol, ether, 
 benzine or a strong acid. If no appreciable residue remain, 
 the fluid is probably pure water, or some other volatile liq- 
 uid; or it is possible that the liquid is some very dilute 
 solution, like a spring water, which needs extreme concen- 
 tration before the solid substances dissolved in it can be 
 detected. When a residue remains on the foil, the heat is 
 increased : first, to ascertain if the dissolved substances are 
 wholly volatile, in which case only compounds of ammo- 
 nium, mercury, arsenic and antimony can be present; and 
 secondly, to ascertain if there be any organic matter in the 
 liquid. Carbonization or charring with the attendant phe- 
 nomena (§ 82, 1) occurs when fixed organic matter is present. 
 If organic matter is discovered, it must be destroyed by 
 the second method of § 84, before the analysis can be pro- 
 ceeded with. A volatile organic solvent can, of course, be 
 got rid of by a simple evaporation to dryness. 
 
 100. Testing with Litmus. — The next step is to test the 
 solution with litmus paper. 
 
 a. If it is neutral, and the solvent is water, consult § 88. 
 
 6. If it is acid, the acidity may be due to a normal salt 
 having an acid reaction, or to an acid salt or to free acid. 
 No general inferences can be drawn from the acid reaction, 
 
 146 
 
146 TBEATMENT OF LIQUIDS. 
 
 except that carbonates and sulphides are absent. If dilu- 
 tion of the acid fluid produces turbidity, the presence of 
 antimony or bismuth may be inferred, 
 c. If it is alkaline, consult § 88, I, c. 
 
 101. By evaporating a portion of the original solution to 
 dryness, the dissolved solid is obtained. This solid may 
 be subjected to the whole of the preliminary treatment 
 prescribed for a salt, mineral or other non-metallic solid 
 (§§ 82, 83) ; but inasmuch as the main object of all prelimi- 
 nary treatment of a solid is to learn how to get it into solu- 
 tion with the least difiiculty, it is seldom worth while for the 
 analyst to make a solid out of a solution, and thus forego 
 the advantage of having the solution already made to his 
 hand. 
 
 102. Testing for Ammonia. — A small portion of the orig- 
 inal solution must always be tested for ammonium salts by 
 heating it in a test-tube with an equal bulk of slaked lime.jrK^ 
 The gas is recognized by its smell and its reaction with 
 hydrochloric acid (§ 82, II, h), <^ '^^rA-' t^A^^^^r^^^'^ \.tjv,^ 
 
 The means of identifying and isolating the rare elements, 
 the methods by which minute traces of one substance may 
 be detected when hidden in proportionally large quantities 
 of other substances, as when the impurities of chemicals and 
 drugs are exhibited, and the processes to be employed in 
 special cases of peculiar difficulty, such as the analysis of 
 complex insoluble minerals, or the detection of mineral poi- 
 sons in masses of organic matter, must be studied in complete 
 treatises upon chemical analysis, or in works specially 
 devoted to these technical matters. Such details, however 
 valuable to the pi-ofessional analyst, or expert, would not 
 be in harmony with the plan of this manual. 
 
APPENDIX. 
 
APPENDIX, 
 
 REAGENTS. 
 
 [Those reagents in the following list which are used in considerable 
 quantities are marked with an asterisk (*).] 
 
 The ordinary commercial substances are, as a rule, sufficiently 
 pure for the purposes of this manual. If in any experiment doubts 
 arise as to the character of a reagent, a quantity of it somewhat 
 larger than that which has been mixed with the substance under ex- 
 amination should be tested by itself, and the reaction compared with 
 that exhibited in the doubtful case. If the result of this trial is unsat- 
 isfactory, the experiment must be repeated with reagents which are 
 known to be pure. 
 
 1. * Hydrochloric Acid (Concentrated). — The strong, com- 
 mon acid prepared by chemical manufacturers, though usually far 
 from pure, will answer for most of the purposes of this manual. It 
 must, however, be continually borne in mind that the commercial acid 
 is usually contaminated with sulphuric acid, and very often with traces 
 of arsenic and iron. These impurities may be present in sufficient 
 quantity to render the acid unfit for use when these very substances 
 are to be tested for in the mixture to be analyzed. 
 
 The yellow color of the commercial acid, though often attributed 
 to iron, is really due for the most part to the presence of a peculiar 
 organic compound which is soluble in the strong acid. 
 
 2. Hydrochloric Acid (Pure). 
 
 3. * Hydrochloric Acid (Dilute). — Mix 1 volume of the com- 
 mon concentrated acid, or — where special purity is required — of the 
 pure strong acid, with 4 volumes of water. 
 
 149 
 
150 BE AGENTS. [§§ 4-13. 
 
 4. * Nitric Acid (Concentrated). — Use the colorless commer- 
 cial acid of 1.38 or 1.40 specific gravity. Strong nitric acid of tolerable 
 purity can be obtained from the dealers in coarse chemicals. An acid 
 which, when diluted with 5 parts of water, gives no decided cloudi- 
 ness with either nitrate of silver (absence of hydrochloric acid) or 
 nitrate of barium (absence of sulphuric acid) is good enough for most 
 uses in qualitative analysis. 
 
 5. Nitric Acid (Pure). 
 
 6. * Nitric Acid (Dilute). — Mix 1 volume of the strong acid with 
 5 volumes of water. 
 
 7. Aqua regia should be prepared only in small quantities, at 
 the moment of use, by mixing in a test-tube 1 volume of strong 
 nitric acid with 3 or 4 times as much strong hydrochloric acid. 
 
 8. * Sulphuric Acid (Concentrated). — The oil of vitriol of 
 commerce will usually be found pure enough for the purposes of this 
 manual. 
 
 9. Sulphuric Acid (Pure). — Sulphuric acid free from hydro- 
 chloric acid is necessary in testing for chlorine, according to § 72. 
 Such acid may be obtained from the dealers in chemicals, but acid 
 purporting to be pure should invariably be tested by diluting a portion 
 with a considerable quantity of water and adding a few drops of 
 nitrate of silver. No turbidity should appear after the mixture has 
 stood for some time. 
 
 10. * Sulphuric Acid (Dilute) is prepared by gradually adding 
 1 part by measure of the concentrated acid to 4 parts of water con- 
 tained in a beaker or porcelain dish ; the mixture must be constantly 
 stirred with a glass rod. When the mixing is finished, the liquid is 
 left at rest until all the sulphate of lead, which has separated from the 
 strong acid, has settled to the bottom ; the clear liquid is then decanted 
 into bottles. 
 
 11. Sulphuric Acid (Dilute). — Mix, as described in the pre- 
 ceding section, 1 part of the strong acid with 6 parts of water. 
 
 12. Oxalic Acid. — Dissolve 1 part, by weight, of the commer- 
 cial crystals in 20 parts of water. 
 
 13. * Acetic Acid. — The ordinary commercial acid ; or dilute 
 Glacial acetic acid with two and a half times its own volume of 
 water. 
 
§§14-18.] BEAGENTS. 151 
 
 14. Tartaric Acid should be kept in the state of powder, since 
 solutions of it slowly decompose. For use, dissolve a small portion of 
 the powder in 2 or 3 times its volume of hot water. 
 
 15. * Sulphuretted Hydrogen Gas (Sulphydric Acid) is pre- 
 pared as needed, by acting upon fragments of sulphide of iron with 
 dilute sulphuric acid in the apparatus described in §§ 93, 94, of this 
 Appendix. The apparatus should always be placed either in the open 
 air or in a strong draught beneath a chimney. 
 
 16. Sulphuretted Hydrogen "Water. — Pass sulphuretted hydro- 
 gen gas into a bottle of water until the water can absorb no more. 
 To determine when the absorption is complete, close the mouth of the 
 bottle tightly with the thumb, and shake the liquid. If the water is 
 saturated, a small portion of the gas will be set free by the agitation, 
 and a slight outward pressure against the thumb will be felt. If the 
 water is not fully saturated, the agitation will enable it to absorb the 
 gas which lay in the upper part of the bottle, and a partial vacuum 
 will be created, so that an inward pressure will be felt. 
 
 Since sulphuretted hydrogen water soon decomposes when exposed 
 to the air, it should always be kept in tightly closed bottles, and no 
 very large quantity of it should be prepared at once. A good way of 
 keeping the solution is to fill a number of small phials with the fresh 
 liquid, cork them tightly, and invert them in water, so that their necks 
 shall always be immersed and protected from the atmosphere. 
 
 At the moment of using this reagent its quality should always be 
 proved by smelling of it, or by adding a drop or two of the liquid to 
 a drop of acetate of lead, which should be immediately blackened 
 from the formation of sulphide of lead. 
 
 17. * Ammonia- Water. — Commercial aqua-ammoniae may usu- 
 ally be obtained pure enough for the purposes of this manual. Dilute 
 1 volume of the strong liquor with 3 volumes of water. Ammonia- 
 water should be free from carbonic acid ; when diluted, as above, it 
 ought not to yield any precipitate when tested with lime-water. 
 
 18. * Sulphide of Ammonium. — Pass sulphuretted hydrogen 
 gas through ammonia-water, diluted as described in § 17, until a por- 
 tion of the liquid yields no precipitate when tested with a drop of a 
 solution of sulphate of magnesium (absence of free ammonia). 
 
 Since sulphide of ammonium decomposes after a while, when 
 exposed to the air, it is not advisable to prepare it in large quantities. 
 In case any doubt arise as to the quality of the reagent, add some of 
 
152 BEAGENTS. [§§ 19-23. 
 
 it to a drop of acetate of lead. Unless a dense "black precipitate 
 of sulphide of lead is immediately thrown down, the sulphide of 
 ammonium is worthless. The sulphide as prepared above, if allowed 
 to stand for some time exposed to light, usually becomes sufficiently 
 transformed to the yellow sulphide for use. But yellow sulphide of 
 ammonium may be prepared directly by dissolving sulphur to satura- 
 tion in the above-mentioned solution. 
 
 19. * Carbonate of Ammonium. — Dissolve without warming 
 1 part, by weight, of the commercial salt, in 4 parts of water, and add 
 to the mixture 1 part of strong ammonia-water. 
 
 The solution of carbonate of ammonium should be kept in bottles 
 made of glass which is free from lead. If kept in flint-glass bottles, 
 the carbonate of ammonium takes up some lead so that when, in pre- 
 cipitating the carbonates of Class VI, this reagent is added to a solu- 
 tion containing sulphide of ammonium, a dark coloration appears in 
 the liquid or obscures the precipitate, to the annoyance of the operator. 
 
 20. * Chloride of Ammonium. — Dissolve without warming 1 
 part, by weight, of the crystallized commercial salt in 10 parts of 
 water ; let the solution stand for a day or so and then filter it. 
 
 21. * Oxalate of Ammonium. — Dissolve 1 part, by weight, of 
 the salt in 24 parts of water ; or, dissolve 37 grms. of crystallized 
 oxalic acid in 450 cubic centimetres of water, neutralize exactly with 
 ammonia-water and dilute to the bulk of 1 litre. 
 
 22. Molybdate of Ammonium. — Dissolve 1 part of molybdic 
 acid in 4 parts of strong ammonia-water, taking care to add only a 
 little molybdic acid at a time, and to stir after each addition. Filter 
 the ammoniacal solution and allow it to cool. Mix the clear solution 
 with 15 parts of strong nitric acid, adding the ammonia solution 
 slowly, with constant stirring, taking care to keep the mixture cool. 
 Let the mixture stand 24 hours in a warm place and filter ; or 
 dissolve 1 part of molybdate of ammonium in 3 or 4 parts of weak 
 ammonia-water, and mix the liquid with 12 or 15 parts of nitric acid, 
 as before. 
 
 23. * Sodium Hydrate. — Place 1 part, by weight, of the best 
 commercial caustic soda in a large, stoppered bottle ; pour upon it 8 
 or 9 parts of water, and shake the bottle at intervals until the whole 
 of the soda has dissolved. Leave the bottle at rest until the liquid 
 has become clear, and finally transfer the solution, with a siphon, to 
 the small bottles in which it is to be kept for use. The solution thus 
 
§§ 24, 25.] REAGENTS. 153 
 
 prepared, though pure enough for the uses prescribed in this manual, 
 is really far from pure. It would be unfit for use in a delicate research, 
 because it is usually contaminated with chloride, sulphate and carbo- 
 nate of sodium, and is liable to contain traces of aluminate, phosphate 
 and silicate of sodium. Since some nitrate of sodium is added to it 
 in the process of manufacture, the soda is liable to be contaminated 
 with this salt and the products of its decomposition, including ammonia. 
 This last impurity is liable to be given off when the solution is boiled. 
 
 Potassium hydrate (caustic potash) may be substituted for caustic 
 soda whenever it can be more readily obtained. The potash should 
 be dissolved in about 10 parts of water. 
 
 Since solutions of the caustic alkalies act upon glass rather easily, 
 especially when its outer surface or "fire-glaze" has once been 
 removed, it often happens, when the soda solution is kept in glass- 
 stoppered bottles, that the stoppers become immovably cemented to 
 the glass by the silicate of sodium which forms in their necks. This 
 difficulty may be avoided by wiping the necks of the bottles dry after 
 any of the solution has been poured from them ; but it will usually be 
 found more convenient to replace the glass stoppers with plugs of 
 vulcanized caoutchouc, or better still, with small glass stoppers, over 
 the bodies of which short pieces of caoutchouc tubing have been 
 stretched. Solutions of the caustic alkalies should be kept in bottles 
 made of glass which is free from lead. 
 
 24. * Sulphide of Sodium. — Dissolve 1 part, by weight, of com- 
 mercial sulphide of sodium, if it can be obtained, in 8 parts of water. 
 Sulphide of sodium may be made by melting together in an iron pot 
 or ladle a mixture of dry carbonate of sodium with an equal weight of 
 sulphur. This operation does not require a great deal of heat, and it 
 may be performed over the blast lamp or over an ordinary coal fire. 
 Sulphide of potassium is not an available substitute for sulphide of 
 sodium. 
 
 25. * Carbonate of Sodium. — For most purposes it is essential 
 that the sodium carbonate should be free from sodium sulphate. 
 Pure dry carbonate may be obtained from the dealers in chemicals, 
 or it may be prepared by washing a pound or two of bicarbonate of 
 sodium repeatedly, upon a filter, with small quantities of ice-cold 
 water, until the original quantity is reduced to a fifth or a sixth of its 
 bulk. The powder is then dried, ignited, and kept in well-stoppered 
 bottles. For the solution, dissolve 1 part of the salt in 4 parts of 
 water. 
 
154 BE AGENTS, C§§ 26^6- 
 
 26. Biborate of Sodium (Borax). — Common borax, powdered. 
 
 27. Phosphate of Sodium. — Dissolve 1 part, by weight, of 
 " common phosphate of soda " in 10 parts of water. 
 
 28. Acetate of Sodium. — Dissolve 1 part of the crystallized salt 
 in 10 parts of water. 
 
 29. Nitrate of Sodium. — Select a clean, white sample of the 
 commercial salt, and keep in the form of a coarse powder. 
 
 30. Acid Sulphate of Sodium. — Heat a mixture of 16 parts, 
 by weight, of Glauber's salt, and 5 parts of concentrated sulphuric 
 acid, in a platinum vessel, until a portion of the melted mass becomes 
 distinctly solid when taken up on a glass rod. Then allow the mix- 
 ture to become cold ; remove the cold lump from the platinum vessel, 
 and break it into fragments. Keep the coarse powder in a tight, glass- 
 stoppered bottle. The commercial salt will answer. 
 
 31. Sulphate of Potassium. — Dissolve 1 part, by weight, of the 
 crystallized salt, in 200 parts of water. A solution of this strength 
 contains the same proportional quantity of sulphuric acid as is con- 
 tained in a saturated aqueous solution of sulphate of calcium. Hence 
 it cannot precipitate the latter when added to solutions of the soluble 
 calcium salts. 
 
 32. Chromate of Potassium. — (The normal or "neutral" yel- 
 low chromate.) Dissolve 1 part, by weight, of the salt in 8 parts of 
 water. 
 
 33. Bichromate of Potassium. — The pure crystallized salt is 
 kept in the form of powder. It must be entirely free from chloride. 
 
 34. Ferrocyanide of Potassium. — ( Yellow Prussiate of Potash.) 
 Dissolve 1 part, by weight, of the commercial salt in 12 parts of 
 water. 
 
 35. Ferricyanide of Potassium. — (Red Prussiate of Potash.) 
 Since the aqueous solution of this salt undergoes decomposition, with 
 formation of some ferrocyanide, when kept for any length of time, 
 the salt should be kept for use in the form of powder. The com- 
 mercial salt is pure enough for analytical purposes. A minute frag- 
 ment of it may be dissolved in water at the moment of use. 
 
 36. Cyanide of Potassium. — The better sorts of the commer- 
 cial article are pure enough for analytical purposes. It should be 
 kept in the solid form in a tightly stoppered bottle. When the solu- 
 tion is required, dissolve 1 part of the salt in 4 parts of cold water. 
 
§§ 37-43] BE AGENTS. 155 
 
 37. Nitrate of Potassium. — Refined saltpetre may be employed. 
 It should be kept in the state of powder. 
 
 38. Nitrite of Potassium. — Weigh out 8 parts of concentrated 
 nitric acid, mix it with an equal weight of water, and place the mix- 
 ture in a glass flask provided with a perforated cork and gas delivery- 
 tube. The flask should be so large that the mixture only half fills it. 
 Throw into the liquid 2 parts of starch, in lumps, and heat the mix- 
 ture imtil red fumes of nitrous and hyponitric acids begin to be given 
 off ; then remove the lamp lest the action become too violent. Conduct 
 the fumes into a bottle containing 5 parts of potash-lye of 1.27 sp. 
 gr., until the latter is saturated. Then filter the saturated liquid, and 
 evaporate it to dryness. For use, dissolve 1 part of the dry salt in 2 
 parts of water. 
 
 The nitrite of potassium bought of dealers in fine chemicals is 
 often unfit for the uses prescribed in this manual ; it can readily be 
 made good, however, by dissolving it in twice its weight of water and 
 saturating the solution with nitrous fumes. 
 
 39. Iodide of Potassium and Starch Papers. — Dissolve a 
 gramme of pure iodide of potassium (free from iodate) in 200 cubic 
 centimetres of water. Heat the solution moderately in a porcelain 
 dish, and stir into it 10 grms. of starch which has been reduced to 
 the consistence of cream by rubbing it in a mortar with a small 
 quantity of water. Stir the mixture until it gelatinizes, taking care 
 not to bum the starch, then allow the paste to cool, and spread it 
 thinly upon one side of white glazed paper with a wooden spatula. 
 Dry the paper, cut it into strips as large as the little finger and pre- 
 serve it in stoppered bottles kept carefully closed. 
 
 40. Nitrate of Silver. — Dissolve 1 part, by weight, of the com- 
 mercial crystals in 20 parts of water. 
 
 41. Slaked Lime. — Mix common quicklime with half its weight 
 of water. Keep the powder in bottles with tight stoppers. 
 
 42. * Lime- Water. — Place a handful of slaked lime in a large 
 bottle, pour in enough water to almost fill the bottle, cork the latter 
 tightly, and shake it at intervals during several days. Decant the 
 clear liquid into smaller bottles for use. Refill the large supply-bottle 
 with water, and again shake it at intervals. 
 
 43. Chloride of Calciimi. — Stir powdered white marble into 
 dilute hydrochloric acid until the acid is saturated, and dilute 1 part 
 of the concentrated solution with 5 parts of water. 
 
156 KEAGENTS. [§§44-53. 
 
 44. * Chloride of Barium. — Dissolve 1 part, by weight, of the 
 commercial salt in 10 parts of water. 
 
 45. Nitrate of Barium. — Dissolve 1 part, by weight, of the 
 commercial salt in 15 parts of water, 
 
 46. * Acetate of Lead. — Dissolve 1 part, by weight, of " sugar 
 of lead " in 10 parts of water, and filter. 
 
 47. Lead Paper. — Slightly moisten filter paper with a solution 
 of acetate of lead at the moment of use. 
 
 48. Chloride of Magnesium and Chloride of Ammonium 
 (Magnesium Solution) . — Dissolve, without heating, 65 grms. of 
 crystallized chloride of magnesium and 165 grms. of commercial chloride 
 of ammonium in water ; add 260 cubic centimetres of ammonia-water, 
 and dilute the liquor to the volume of a litre. If less than a litre of the 
 reagent is required, the weights above given may, of course, be reduced 
 in any desired proportion. Let the solution stand two or three days, 
 and filter it, to separate any precipitate of ferric hydrate or other 
 insoluble matters, which may have been present as impurities in the 
 components of the mixture. Preserve the clear liquid. 
 
 From a solution thus prepared no hydrate of magnesium can be 
 precipitated by ammonia-water ; herein consists the advantage of the 
 mixture as a test for phosphoric and arsenic acids, 
 
 49. Ferric Chloride. — Dissolve 1 part, by weight, of the salt in 
 15 parts of water ; or this solution can be prepared by passing chlorine 
 gas through a saturated solution of iron tacks in hydrochloric acid, 
 until a drop of the fluid no longer produces a blue precipitate in a 
 solution of ferricyanide of potassium. The solution is then heated, to 
 expel the excess of chlorine. 
 
 50. Nitrate of Cobalt. — Dissolve 1 part, by weight, of the crys- 
 tallized salt in 10 parts of water. 
 
 51. Sulphate of Copper. — Dissolve 1 part, by weight, of the 
 crystallized salt (blue vitriol) in 10 parts of water. 
 
 52. Stannous Chloride. — This solution is prepared by boiling 
 scraps of tin with strong hydrochloric acid until hydrogen ceases to be 
 evolved. The tin must be in excess. The solution is diluted with 
 4 times its bulk of water acidulated with hydrochloric acid, and 
 filtered, if necessary. The clear liquid must be kept in a tightly closed 
 bottle containing some bits of tin. 
 
 53. Black Oxide of Manganese. — The artificially prepared 
 pure binoxide of manganese. 
 
§§ 54-59.] REAGENTS. 157 
 
 54. Red Ozide of Mercury. — The commercial oxide. It should 
 leave no residue when heated upon platinum foil. 
 
 55. Chloride of Mercury. — Dissolve 1 part of mercuric chloride 
 (" corrosive sublimate ") in 20 parts of water. 
 
 56. Platinic Chloride. — Dissolve 1 part of the salt in 10 parts 
 of water. Also prepared by dissolving old platinum foil in aqua regia 
 as follows : cut the worn-out platinum foil and scraps of wire into 
 very fine pieces, and boil them in a porcelain dish, with successive 
 small portions of aqua regia, until all the metal has dissolved. Col- 
 lect the several portions of aqua regia partially saturated with plati- 
 num in another dish, and evaporate nearly to dryness on a water- bath. 
 Dissolve the residue in water, with the addition of a few drops of 
 hydrochloric acid, and filter. Add chloride of ammonium solution to 
 the clear liquid ; collect the precipitated chloroplatinate of ammonium 
 on a filter, wash thoroughly, dry and ignite in a porcelain crucible or 
 small evaporating-dish. Redissolve the spongy platinum obtained in 
 aqua regia. Evaporate to dryness on a water-bath, adding a little 
 hydrochloric acid. Dissolve the residue in 10 parts of water. 
 
 57. Zinc. — The commercial sheet metal, although usually con- 
 taminated with lead and cadmium, and often containing faint traces 
 of arsenic and sulphur, will generally be found pure enough for the 
 purposes of this manual. 
 
 58. "Solution of Indigo" (Sulphindigotic Acid).— Pour 5 
 
 parts (5 grms. will be ample) of fuming sulphuric acid into a beaker, 
 place the latter in a dish of water to keep it cool, and stir into 
 the acid, little by little, 1 part of finely powdered indigo. When all 
 the indigo has been added to the acid, leave the mixture at rest for 
 48 hours ; then pour it into 20 times its own volume of water, filter 
 the mixture, and preserve the filtrate for use. Instead of 6 parts of 
 fuming sulphuric acid, 12 or 14 parts of the ordinary strong acid may 
 be employed ; in this case, however, the mixture must be heated for 
 several hours on a water-bath. The commercial " Indigo Paste " will 
 answer. 
 
 59. Litmus Paper. — Heat 1 part, by weight, of commercial litmus 
 with 6 parts of water, upon a water- bath for several hours, taking 
 care to replace the water which evaporates. Filter, divide the filtrate 
 into two equal portions, and stir one half repeatedly with a glass rod 
 dipped in very dilute nitric acid, until the color appears distinctly red. 
 Pour the blue and red halves into a porcelain dish, and stir the mix- 
 ture. Draw strips of fine unsized paper through the liquid, and hang 
 
158 REAGENTS, [§§ 60-66. 
 
 them on cords to dry. The color of the paper thus obtained is not 
 blue, but bluish- violet. It turns blue when touched with an alkali, 
 and red when exposed to acids, and may be used indifferently as a 
 test for either acids or alkalies. 
 
 60. Starch Paste should be prepared, when wanted for use, by 
 boiling 30 cubic centimeters of water in a porcelain dish, and stirring 
 into it half a gramme of starch which has previously been reduced to 
 the consistence of cream by rubbing it in a mortar with a few drops 
 of water. 
 
 61. Alcohol. — Common alcohol of 85 or 90 per cent. 
 
 62. Water. — Clean rain-water will serve well enough for most 
 of the purposes of this manual. In granitic regions the water of many 
 lakes, brooks and ponds also is nearly pure. Pure water may be 
 obtained by melting blocks of compact ice, or by distilling ordinary 
 water in glass or copper retorts and rejecting the first portions of the 
 distillate. It should yield no precipitate when tested with chloride of 
 barium and nitrate of silver. 
 
 63. Hypochlorite of Sodium. — The "chloride of soda" of the 
 druggists. Prepare by shaking up 1 part of bleaching powder with 10 
 parts of water ; add a saturated solution of commercial carbonate of 
 sodium as long as a precipitate is produced. Let the turbid mixture 
 settle, and siphon off the clear liquid. 
 
 64. Bisulphide of Carbon. — Commercial. Great care should be 
 taken in using the reagent, as it is extremely inflammable, and the 
 vapor mixed with air is violently explosive. 
 
 65. Chlorine Water. — A solution of chlorine in water. It should 
 be kept in a well-stoppered bottle and not exposed to light. Otherwise 
 it is speedily converted into hydrochloric acid with evolution of 
 oxygen. 
 
§66.] KNOWN SOLUTIONS. 159 
 
 SOLUTIONS OF KNOWN COMPOSITION. 
 
 66. In case the experiments indicated in Part I are to be per- 
 formed by a considerable number of students, it will be found conven- 
 ient to prepare beforehand a moderate supply of the various solutions 
 required in making the known mixtures under the several classes. 
 These solutions may be made of the strengths indicated below. 
 
 Chloride of Copper. — Dissolve black oxide of copper in 5 times 
 its weight of a mixture of equal parts of strong hydrochloric acid and 
 water. Dilute the resulting solution with 3 times its bulk of water. 
 [A single student, in performing the experiment, may dissolve a few 
 grains of the oxide in a small quantity of the strong acid, and, in 
 general, may make the solutions as needed for use by taking a crystal 
 or a small amount of the required substance in powder, as the case 
 may be, without regard to the exact amount. It is well, however, not 
 to start with such quantities as to make the precipitates inconveniently 
 bulky.] 
 
 Araenious Oxide. — Dilute a quantity of hydrochloric acid with 
 half its bulk of water, and saturate it with arsenious oxide at a gentle 
 heat. When the solution has become cold, pour off the clear liquor 
 from the arsenious oxide which has crystallized out. 
 
 Ferrous Chloride. — Treat warm dilute hydrochloric acid (App., 
 § 3) with as much iron (wire or filings) as it will dissolve, and then 
 dilute the solution with an equal bulk of water. [This solution should 
 be prepared only in small quantity, and kept in a well-stoppered 
 bottle.] 
 
 Chloride of Zinc*. — To a quantity of hydrochloric acid diluted 
 with an equal bulk of water, add as much zinc as the acid will dis- 
 solve, and then add to the solution 5 times its bulk of water. 
 
 Chloride of Calcium. — Stir powdered white marble or chalk into 
 hydrochloric acid diluted with twice its bulk of water until the acid is 
 saturated ; filter the solution, if necessary. 
 
 Chloride of Magnesium. — Dissolve 1 part of the salt in 10 parts 
 of water, or add " magnesia alba " to dilute hydrochloric acid until 
 the acid is saturated, then dilute the solution with twice its bulk of 
 water. 
 
 Chloride of Sodium. — Dissolve common salt in 10 times its 
 weight of water. 
 
160 KNOWN SOLUTIONS. [§66. 
 
 Nitrate of Silver. — Dissolve the crystallized salt in 10 times its 
 weight of water. 
 
 Mercurous Nitrate. — Dilute a small quantity of strong nitric 
 acid with an equal bulk of water, and to the mixture, warmed over 
 the lamp, add more mercury than will dissolve. When action has 
 ceased, dilute the solution with 5 times its bulk of water and keep in 
 a bottle containing a small amount of metallic mercury. 
 
 Nitrate of Lead. — Dissolve the crystallized salt in 5 times its 
 weight of water. 
 
 Mercuric Chloride. — Dissolve corrosive sublimate in 20 times its 
 weight of water. 
 
 Chloride of Bismuth. — Dissolve metallic bismuth in aqua regia. 
 When the acid is saturated pour off the solution from the undissolved 
 metal, dilute it with twice its bulk of water and add strong hydro- 
 chloric acid to dissolve the precipitated oxy- chloride. Or, dissolve the 
 commercial sub-nitrate in hydrochloric acid and dilute as before. 
 
 Chloride of Cadmium. — Dissolve the commercial salt in 10 times 
 its weight of water. 
 
 Chloride of Lead. — Boil dilute hydrochloric acid with an excess 
 of litharge and filter the solution when perfectly cold. 
 
 Chloride of Antimony. — Dilute the commercial, strong solution 
 with an equal bulk of water, and add strong hydrochloric acid to dis- 
 solve the basic chloride which is precipitated. Or, dissolve the finely 
 powdered metal in aqua regia and dilute as before. 
 
 Sulphate of Manganese. — Dissolve the crystallized salt in 10 
 times its weight of water. 
 
 Common Alum. — Dissolve in 10 times its weight of water. 
 
 Chrome Alum. — Dissolve in 10 times its weight of water without 
 heating. 
 
 Bone Ash (p. 47) had better be kept in powder and dissolved as 
 needed, in order that the student may not lose sight of the fact that 
 this compound requires an acid solvent. 
 
 Nitrate of Cobalt. — Dissolve in 10 times its weight of water. 
 
 Nitrate of Nickel. — Dissolve in 10 times its weight of water. 
 (The chlorides of nickel and cobalt answer equally well, and the solu- 
 tions may be made of the same strength.) 
 
^66.] KNOWN SOLUTIONS. 161 
 
 Chloride of Barium. — Dissolve the crystallized salt in 5 times 
 its weight of water. 
 
 Chloride of Strontium. — Dissolve the crystallized salt in 5 times 
 its weight of water, (The nitrate will answer equally well.) 
 
 Chloride of Tin. — Use the reagent (App., § 52). 
 
 Nitrate of Potassium. — Dissolve 1 part of the commercial salt 
 in 5 parts of water. 
 
 Sulphate of Sodiiun. — Dissolve 1 part of Glauber's salt in 10 
 parts of water. 
 
 Phosphate of Sodium. — Dissolve commercial ''phosphate of 
 soda " in 10 parts of water, as in App., § 27. 
 
 Carbonate of Sodium. -— Dissolve 1 part of "sal soda" in 5 
 parts of water. 
 
 Oxalate of Potassium. — Dissolve 1 part of the crystallized salt 
 in 5 parts of water. 
 
 Tartrate of Potassium. — Dissolve tartaric acid in 5 times its 
 weight of water and neutralize it exactly with carbonate of potassium. 
 
 Iodide of Potassium. — Dissolve 1 part of the crystallized salt in 
 10 parts of water. 
 
162 
 
 tlTHNSlLS. 
 
 in 67, 68. 
 
 UTENSILS. 
 
 67. The Implements required by the student of qualitative 
 analysis are few and simple. Besides bottles for the reagents enu- 
 merated in the foregoing list, and a few small phials for the preser- 
 vation of samples of salts and mixtures to be analyzed, there will 
 be needed — 
 
 A dozen test-tubes, 
 A wooden test-tube rack, 
 A test-tube brush, 
 A nest of small beakers, 
 2 or 3 glass stirring-rods, 
 A small thistle-, or funnel-tube, 
 A larger thistle-tube for the gas- 
 generator, 
 
 1 stick of No. 7 glass tubing (see 
 App., § 86), 
 
 2 or 3 sticks of No. 5 glass tub- 
 ing, 
 
 3 small glass funnels, 
 A small glass flask, 
 
 A small platinum crucible is also 
 
 very desirable, 
 A few packages of cut filters, or a 
 
 quire of filter-paper, 
 
 A wash-bottle, 
 
 2 small evaporating-dishes, 
 
 A porcelain crucible, 
 
 1 triangle of iron wire, 
 
 An iron ring-stand, 
 
 A filter-stand, 
 
 A lamp, 
 
 A gas- bottle for generating sul- 
 phuretted hydrogen, 
 
 A common jeweller's blowpipe, 
 
 A pair of small iron pincers (jew- 
 eller's tweezers), 
 
 A piece of platinum foil, 
 
 A bit of platinum wire, 
 
 A few corks or caoutchouc stop- 
 pers, 
 
 A piece of blue cobalt glass (see 
 §43). 
 
 68. Reagent Bottles. — The bottles in which reagents are kept 
 should be of cylindrical shape, and rather high than wide. They 
 should be closed with glass stoppers which fit accurately, but are not 
 very finely ground. Most of the liquid reagents may be conveniently 
 kept in narrow-mouthed bottles of the capacity of 6 fluid ounces ; but 
 to avoid the necessity of frequently refilling the bottles, it is well to 
 keep the solutions most commonly employed — namely, dilute hydro- 
 chloric and nitric acids, ammonia-water, chloride of ammonium and 
 carbonate of ammonium — in 8- ounce bottles. Care must be taken in 
 this case to choose bottles of such shape that they can be readily 
 grasped between the thumb and fingers. 
 
 For the reagents which are to be kept in the dry state, wide- 
 mouthed bottles of the capacity of 2 or 3 ounces should be chosen. 
 
§69.] BEAGENT BOTTLES. 163 
 
 Reagent bottles should always be made " extra-heavy," since, from 
 constant use, they are exposed to many blows. The lustrous "flint- 
 glass" bottles of American or English make are ill suited for the 
 preservation of liquid reagents ; for such glass is easily attacked by 
 many chemical agents, and is therefore likely to render the reagents 
 impure. The lettered reagent bottles recently introduced and manu- 
 factured by Whitehall, Tatum & Co., Philadelphia, are excellent in 
 every respect. 
 
 Each reagent bottle should be kept in a particular place on shelves 
 before the operator and convenient to his hand. Whenever a reagent 
 is to be used, the bottle which contains it should be grasped in the 
 right hand ; the stopper should be taken out by pinching it between 
 the first and second or third and fourth fingers of the left hand, or by 
 pressing it between the little finger and palm of that hand. In either 
 case, the bottle is withdrawn from the stopper, and not the stopper 
 from the bottle. Neither bottle nor stopper should be put upon the 
 table ; the stopper should be held in the left hand as long as the bottle 
 is open. When the reagent has been poured out, the bottle is imme- 
 diately closed, and returned to its place upon the shelf. If these 
 apparently trifling particulars are scrupulously attended to, no stopper 
 can ever be misplaced, or soiled by contact with liquids or dirt on the 
 table ; and the bottle will always be found in its proper place when 
 Instinctively reached for. Moreover, the label on the bottle cannot be 
 injured by drops of the reagent, since the liquid must necessarily be 
 poured from the back, or blank side of the bottle. 
 
 When a stopper sticks tightly in the neck of a bottle, it may some- 
 times be loosened by pressing it first upon one side, and then upon the 
 other, with the thumb of the right hand, while the fingers of that hand 
 grip the bottle, and the bottle is held still with the left hand. Or the 
 neck of the bottle may be immersed in hot water for a minute or two, 
 to expand the glass outside the stopper. The stopper can then usu- 
 ally be taken out without trouble. The hot water may be conveniently 
 applied by pouring a slow stream of it from a wash-bottle upon the 
 neck of the bottle. Another way is to heat the neck of the bottle 
 over a very small flame of the gas- or alcohol-lamp. No matter how 
 the glass is heated, the bottle must be constantly turned round and 
 round, in order that each side of the neck may be equally exposed to 
 the heat and the risk of cracking the bottle so be lessened. 
 
 69. Test-tubes are little cylinders of thin glass with round, thin 
 bottoms and lips slightly flared. Their length may be from 5 to 7 
 inches, and their diameter from one half to three fourths of an inch ; 
 
164 
 
 TEST-TUBE RACK. 
 
 [§§ 70, 71. 
 
 they should never be so wide that the open end cannot be closed by 
 the ball of the thumb. 
 
 Test-tubes are used for heating small quantities of liquid over the 
 gas- or spirit-lamp ; they may generally be held by the upper end in 
 the fingers without inconvenience ; but in case they become too hot 
 to be held in this way, a strip of thick, folded paper may be nipped 
 round the tube, and grasped between the thumb and forefinger just 
 outside the tube. 
 
 Two precautions are invariably to be observed in heating test- 
 tubes : — 1st. The outside of the tube must be wiped perfectly dry ; 
 and 2d. The tube must be moved in and out of the flame for a minute 
 or two when first heated. It should be rolled to and fro also to a 
 slight extent between the thumb and forefinger, in order that each side 
 of it may be equally exposed to the flame. A drop of water on the 
 outside of the tube keeps one spot cooler than the rest. The tube 
 breaks, because its parts, being unequally heated, expand unequally, 
 and tear apart. 
 
 When a liquid is boiling actively in a test-tube, it sometimes hap- 
 pens that portions of the hot liquid are projected out of the tube with 
 some force ; the tube should therefore always be held in an inclined 
 position, and the operator should be careful not to direct it towards 
 himself, or towards any other person in his neighborhood. 
 
 Test-tubes are cleaned by the aid of cylindrical brushes made of 
 bristles caught between twisted wires, like those used for cleaning 
 lamp-chimneys ; the brushes should have a round end of bristles. 
 
 70. Test-Tube Rack. — Test-tubes are kept in a wooden rack, 
 such as is represented in Fig. 1. When in use, the tubes stand 
 
 upright in the holes of the rack; but 
 clean tubes are inverted upon the pegs 
 behind the holes, in order that they may 
 be kept free from dust, and that the last 
 portions of wash-water may drain away 
 from them after washing. The rack should 
 be large enough to hold a dozen tubes. 
 Care should be taken that the tubes are 
 washed perfectly clean before being in- 
 verted on the pegs, lest the pegs themselves become dirty. 
 
 71. Flasks. — Small flasks of 2 or 3 ounces' capacity are well 
 suited for the purposes of qualitative analysis. When a liquid is to 
 be boiled in a flask, the flask should be placed upon a support of 
 
 Pigr. 1. 
 
§§ 72-74.] FILTERING. 165 
 
 wire-gauze (App., § 80), and sufficiently inclined to prevent any par- 
 ticles of the liquid from being thrown out of the neck by the move- 
 ment of ebullition. 
 
 As with test-tubes and all other glass or porcelain vessels of what- 
 ever form, the outside of a flask nmst be wiped perfectly dry before 
 exposing it to the lamp. The flame should be moved about also 
 beneath the flask, at first, in order that the temperature of the latter 
 may be raised equally and not too rapidly. 
 
 72. Beakers are thin, flat-bottomed tumblers with a slightly 
 flaring rim. They are bought in sets or nests. A nest in which 
 the largest-sized beaker has a capacity of about 6 ounces will be suffi- 
 cient for the requirements of this work. 
 
 73. Glass Funnels should be thin and light, and should be about 
 2 or 2.5 inches in diameter. Their sides should incline at an angle 
 of 60°. The wider the throat of the funnel, the better. 
 
 74. Filtering. — Paper filters are employed in qualitative analysis 
 to separate precipitates from the Uquids in which they have been 
 formed. A good filtering-paper must be porous enough to filter rap- 
 idly, and yet sufficiently close in texture to retain the finest powders ; 
 and it must also be strong enough to bear, when wet, the pressure of 
 the liquid which is poured in upon it. Filter-paper should never con- 
 tain any gypsum or other soluble material, and should leave only a 
 small proportion of ash when burned. White or light-gray paper is to 
 be preferred to colored, since it more commonly fulfils these require- 
 ments. 
 
 Filtering-paper is commonly sold in sheets, which may be cut into 
 circles of any desired diameters for use, according to the various 
 scales of operation, and quantities of liquids to be filtered. Or pack- 
 ages of "cut-filters" may be procured ready-made from the dealers 
 in chemical wares. 
 
 As a general rule, small filters should be employed in analytical 
 operations ; the mixture to be filtered should be poured by small suc- 
 cessive portions upon a filter no larger than is needed to hold the 
 whole of the solid matter which is to be collected. Filters about 3 
 inches in diameter are well suited for most of the analytical operations 
 described in this work, though there are many cases where smaller 
 filters are required, and a few instances in which filters as large as 
 4 inches in diameter might be necessary. 
 
 There are two ready methods of preparing filters for use. According 
 
166 FILTERING. [§ 75. 
 
 to the first method, shown in Fig. 2, a circle of paper is folded over 
 on its own diameter, and the semicircle thus obtained is folded once 
 upon itself into the form of a quadrant ; the paper 
 Fiff- 2. thus folded is opened so that three thicknesses shall 
 
 come upon one side, and one thickness upon the 
 other, as shown in the upper half of Fig. 2 ; the 
 filter is then placed in a glass funnel, the angle 
 of which should be precisely that of the opened 
 paper, viz. 60°. The paper may be so folded as 
 to fit a funnel whose angle is more or less than 
 60°, but this is the most advantageous angle, and 
 funnels should be selected with reference to their 
 correctness in this respect. 
 In the second method of folding filters, the circle of paper is doubled 
 once upon itself as before into the form of a semicircle, and a fold 
 equal to one quarter of this semicircle is turned down on each side of 
 the paper. Each of the quarter semicircles is then folded back upon 
 itself, as shown in the lower half of Fig. 3 ; the filter is opened, with- 
 out disturbing the folded portions, and placed in the funnel. Filtra- 
 tion can be rapidly effected with this kind of filters, for the projecting 
 folds keep open passages between the filter and the 
 ^* * funnel, and thus facilitate the passage of the liquid. 
 
 That portion of the circle of paper, which must 
 necessarily be folded up in order to give the requi- 
 site conical form to a paper filter retards filtration 
 in the first manner of folding, but helps it in the 
 second. 
 
 A filter should always be moistened with water 
 after it has been placed in the funnel, in order 
 that the fibres of the paper may be swollen and 
 the size of its pores diminished, before any of the matter to be filtered 
 can pass into them. 
 
 Coarse and rapid filtration — as in the preparation of reagents — 
 can be effected with paper filters of large size, or with cloth bags ; 
 also by plugging the neck of a funnel or leg of a siphon loosely with 
 tow or cotton. If a very acid or very caustic liquid, which would de- 
 stroy paper, cotton, tow, or wool, is to be filtered, the best substances 
 wherewith to plug the neck of the funnel are asbestos, gun-cotton and 
 glass-wool, neither of which is attacked by such corrosive liquids. 
 
 75. Filter-Stand. — Filters less than 2 inches in diameter may 
 be placed directly in the mouth of a test-tube without need of even a 
 
§760 
 
 FILTRATION. 
 
 167 
 
 Fig. 4. 
 
 funnel to support them ; and in general the funnel which holds a filter 
 may be thrust directly into the mouth of a test-tube whenever the 
 quantity of liquid to be filtered is small, if only an ample 
 exit be provided for the air in the tube, in the manner 
 shown with the bottle of Fig. 4, 
 
 But when the quantity of liquid to be filtered is compar- 
 atively large, or the operations to which the filtrate is to 
 be subjected require that it should be collected in a beaker 
 or porcelain dish, the funnel should have an independent 
 support. The iron ring-stand, described in § 80 of this 
 Appendix, may be used for this purpose in case of need ; 
 but wooden stands of the form represented in Fig. 5, 
 adapted expressly for holding funnels, are very convenient 
 
 and not expensive. The 
 
 Figr. 5. 
 
 
 \ 
 
 Figr. 6. 
 
 horizontal 
 bar which holds the funnel may be 
 fixed at any height on the vertical 
 square rod by means of a wedge- 
 shaped key, whose form is shown 
 in the figure, or by a wooden screw. 
 A fine-grained wood, which does not 
 swell or shrink much, is desirable for 
 filter-stands. 
 
 In general, care should be taken 
 that the lower end of the funnel 
 touch the side or edge of the vessel 
 into which the filtrate de- 
 scends, in order that the 
 liquid may not fall in 
 drops, but run quietly 
 without splashing. 
 76. Rapid Filtration. — Since in the course of an 
 analysis much time is consumed in the process of filtra- 
 tion, it is desirable that this operation should be made as 
 rapid as possible. A considerable advantage over the ordi- 
 nary method may be gained by increasing the length of 
 the tube of the funnel by the addition of a piece of glass 
 tubing a metre or so in length and bent as represented 
 in Fig. 6. When the funnel and the tube are filled with 
 liquid, the difference of pressure on the upper and lower 
 surfaces is great enough to cause a very sensible increase in the rapid- 
 ity of the filtration. A far more efficient method of hastening the 
 
168 
 
 FILTBATION. 
 
 [§76. 
 
 process of filtration by causing a difference in pressure on the upper 
 and lower surfaces of the liquid to be filtered, may be made available 
 wherever a constant supply of water with a fall of 8 or 10 feet can 
 be obtained. The details of the process and of a convenient form 
 of the apparatus which may be employed, will be described presently ; 
 the principle of the method is as follows : — 
 
 The filtrate, instead of being received in a beaker as is usual, is 
 received in a flask from which the air is more or less completely 
 exhausted. This exhaustion is accomplished by the use of a sort 
 of "water-pump" which is an adaptation of a very simple prin- 
 
 Fig. 7. 
 
 ciple. Let ab and cd be two tubes, arranged as 
 represented in Fig. 7. If water be allowed to flow in a 
 constant stream down the tube ab and the amount of 
 water supplied be properly regulated, that part of the 
 tube ab which is below the junction with cd will be 
 filled with bubbles of air, which is drawn in continuously 
 through cd and dragged down by the falling water; if 
 the tube at c be connected with a closed vessel, the air 
 in the vessel will be gradually exhausted. The efficiency of such a 
 pump depends in a measure upon the relative size of the tubes and 
 the amount of water supplied. Various forms of apparatus in which 
 advantage is taken of this general principle might be and have been 
 devised. As adapted for purposes of filtration the apparatus is known 
 as Bunsen's " filter- pump." On account of the great advantage to be 
 gained by its use, the apparatus, and the method of conducting the 
 filtration will be given in detail. 
 
 A strong glass flask (for the purposes of qualitative analysis, one 
 of from 2 to 4 ounces' capacity will answer) is furnished with a doubly- 
 perforated caoutchouc stopper: through one of the perforations is 
 thrust the neck of a glass funnel and through the other a piece of 
 _. No. 7 glass tubing, bent at a right angle. This 
 
 tube serves to connect the flask with the apparatus 
 designed to effect the rarefaction of the air in the 
 flask. If a paper filter were put in the funnel, in 
 the usual way, and filled with liquid, and any con- 
 siderable difference of pressure were to be brought 
 about between the upper and lower surfaces of the 
 liquid in the filter, the paper would be apt to break 
 and let the precipitate fall into the flask. This 
 danger may be avoided by choosing a smooth glass funnel which has 
 an angle as near 60° as possible, and fitting into it a second funnel, or, 
 
§76.] 
 
 FILTRATION. 
 
 169 
 
 rather, a cone of thin platinum foil, the sides of which possess exactly 
 the same inclination as those of the glass funnel. This platinum cone 
 is made by cutting out from a piece of the thinnest platinum foil that 
 can be obtained, a portion of a circle as represented in Fig. 9, a. For 
 use with a funnel 2 inches in diameter, this circle may conveniently 
 have a radius of 1 inch. The foil is then laid upon a piece _ 
 of hard wood and a number of small holes are punched out 
 of it. A ready implement for this purpose is made by 
 grinding off squarely the point of a common sewing-needle 
 and fitting the needle to a wooden handle. When gently 
 tapped with a hammer, this punch forces out a small round 
 bit of the foil and, by subsequently rubbing the foil on the 
 bottom of a mortar with the pestle, any inequalities of 
 surface are avoided. When the foil has been annealed by 
 being heated to redness in the lamp and allowed to cool, it 
 is bent up into the shape of a cone as represented in Fig. 9, b. 
 This cone should have the same angle of inclination as the 
 funnel in which it is to be used, and it is desirable that the funnel 
 should be of an angle of 60° ; still, if the funnel be regular in shape, 
 it can be used although it varies somewhat from this angle. The cone 
 is best fitted to the funnel by the following manipulation : — 
 
 A solid cone of close-grained hard wood, or better of brass (Fig. 
 10, a), which has been turned to an angle of 60°, is laid upon the 
 platinum foil in such a position that the apex of 
 the cone comes at the centre of the circle of which 
 the foil forms a part ; the foil is then folded up 
 and shaped with the fingers so that it fits the cone 
 closely. This wooden or brass cone is not essen- 
 tial ; the platinum cone could be shaped in the 
 funnel with proper care ; it is, however, very con- 
 venient, especially in a laboratory where there are 
 a number of students. In procuring such a cone, 
 it is well to lay out with a protractor on a piece of thin sheet tin an 
 angle of 60°, to cut this out, and then to give the pattern (templet) to 
 the workman employed. The two edges of the foil lap ; and if they 
 were soldered in this position the cone would, of course, have an 
 angle of 60°. The platinum cone is now inserted in the funnel to be 
 used, and opened out a little with the fingers, if necessary, so that it 
 fits the glass. The funnel should differ so little from 60° that the 
 edges overlap each other only slightly more or slightly less than when 
 the foil was fitted to the wooden cone. By means of a sharp point 
 
170 FILTRATION. [§ 76. 
 
 make two scratches on the platinum cone, to show where the over- 
 lapping edges come when the foil is in position, and then remove the 
 cone from the funnel. Hold the platinum by means of a pair of 
 pincers in the same position that it occupied when in the funnel, which 
 is easily done by observing the scratches made. Then heat the cone 
 (Fig. 9, 6) at the outer overlapping edge, with a blowpipe-flame, put 
 on a small amount of powdered borax, heat again, then put on a bit 
 of pure gold, and heat until the gold melts and solders the two edges 
 together. The cone should be held, the gold put on, and the blowpipe- 
 flame directed in such a manner that the melted gold will run down 
 towards the apex of the cone and not in the contrary direction. After 
 dissolving off the adhering borax with warm water, the cone is ready 
 for use. An ordinary paper filter, folded according to the first method 
 of App., § 74, is introduced into this compound funnel in the usual 
 manner ; when carefully moistened and so adjusted that no air-bubbles 
 are visible between it and the glass, this filter, when filled with a 
 liquid, will support the pressure ev6n of an extra atmosphere without 
 breaking. 
 
 A convenient form of the Bunsen pump is represented in Fig. 11. 
 The tubes (a&, 5c, hh) which are represented in the figure may be 
 very conveniently made of quarter-inch lead pipe (see also, p. 172, 
 near bottom) , and the bulb-like enlargement at d may also be made 
 of lead and soldered on to the smaller pipe. The water is supplied in 
 not too large amount by the cock a, and as it passes along the pipe 
 ahc and into the waste ck, it carries with it a continual stream of 
 air-bubbles dragged through hd. This tube, hd, communicates indi- 
 rectly with the flask F, which is not connected immediately with the 
 pump ; the connection is interrupted by the bottle E, furnished with 
 a perforated stopper through which pass three glass tubes. One of 
 these tubes connects with the flask F; one (by means of caoutchouc 
 tubing) with the lead pipe h ; one with the manometer m. The object 
 of this bottle E is to prevent the flow of water into the flask F, in 
 case, as sometimes happens, the operator in letting on too rapid a 
 stream of water causes it to rise in the bulb d and flow over in the 
 tube hh. All the connectors should be of very thick caoutchouc 
 tubing tightly fitted and then varnished with a strong solution of 
 shellac; the stopper of the bottle E may be treated in the same 
 manner. 
 
 The manometer bottle G is a, convenient but not absolutely essential 
 addition. It consists simply of a small bottle containing a quantity of 
 mercury, to the surface of which the atmosphere has access through a 
 
76.] 
 
 FILTRATION. 
 
 171 
 
1T2 
 
 PORCELAIN DISHES. 
 
 [§77. 
 
 small glass tube. The tube m dips below the mercury and connects 
 with the bottle E. As the air is rarefied in E (and in the flask F), 
 the mercury rises in the tube m ; this tube may be marked in centi- 
 metre or inch divisions by means of a file, or it may be provided with, 
 or attached to, a paper or wooden scale. 
 
 The amount of rarefaction that can be produced by this means 
 depends, of course, upon the head of water accessible, and this will be 
 determined by the difference in height between the points b and k, 
 the extremity of the waste-pipe. With a fall of 35 feet it is possible to 
 obtain nearly a perfect vacuum, but a fall of 8 or 10 feet is sufficient 
 for the purposes of qualitative analysis. 
 
 In operating the filter-pump, certain precautions should be observed. 
 Care should be taken that the water be not supplied in too large an 
 amount. To this end it is well to have the cock or valve so arranged 
 as to make it impossible to let on more water than 
 experience shows to be necessary. The greatest effect 
 is produced by the smallest amount of water that will 
 drag the air down the pipe ck in the form of bub- 
 bles, and in order to observe the flow of air it is well 
 to make a portion of the pipe k of glass. (For that 
 matter, all the pipes may be made of glass tubes 
 joined by thick caoutchouc connectors ; glass tubing 
 of size No. 5 (see App,, § 86) will answer; and the 
 bulb d may also be blown of glass by a person pos- 
 sessing sufficient skill.) In using the pump, the water 
 should be let on before the connection is made with 
 the filtering flask, and the flask should be removed 
 before the water is shut off. 
 
 Instead of Bunsen's pump, a small "Eichards's 
 Aspirator" (Fig. 12), may be used where the water 
 supply is abundant. This " filter-pump " is screwed 
 to a faucet. Its mode of action will appear from the 
 diagram (Fig. 12). It can be got from dealers in 
 chemical supplies. 
 
 77. Porcelain Dishes and Crucibles. — Small 
 open dishes which will bear lieat without cracking, 
 are much used for boiling and evaporating liquids. 
 The best evaporating-dishes are those made of Berlin 
 porcelain, glazed both inside and out, and provided 
 with a little lip projecting beyond the rim. The dishes made of 
 Meissen porcelain are not glazed on the outside, and are not so durable 
 
 Air 
 
 Foam 
 
§78.] LAMPS. . 173 
 
 as those of Berlin manufacture ; but they are much cheaper, and with 
 proper care last a long time. 
 
 The small Berlin dishes, Nos. "00," "0," and " 1," are well suited 
 for all the requirements of this work. They will generally bear an 
 evaporation to dryness and subsequent ignition over the open flame of 
 a gas-lamp, — as when ammonium salts are expelled from Class VII 
 (§ 43), — but it is well to protect the dish somewhat by placing it 
 upon a piece of wire-gauze, rather than to support it upon a simple 
 triangle. The Meissen dishes do not so well endure an open flame. 
 The cheaper kinds of evaporating-dishes, made of "semi-porcelain," 
 should never be subjected to this severe treatment ; they are, for that 
 matter, unfit for use in qualitative analysis. 
 
 Very thin, highly glazed porcelain crucibles, with glazed covers, 
 are made both at Berlin, and at Meissen near Dresden. In general 
 the Meissen crucibles are thinner than the Berlin, but the Berlin 
 crucibles are somewhat less liable to crack ; both kinds are glazed 
 inside and out, except on the outside of the bottom. The Berlin Nos. 
 "00" and "0," respectively 1\ and 1^ inches in diameter, are best 
 suited for the purposes of this manual. As the covers are much less 
 liable to be broken than the crucibles, it is advantageous to buy more 
 crucibles than covers, whenever it is possible so to do. Porcelain 
 crucibles are supported over the lamp on an iron-wire triangle ; they 
 must always be gradually heated, and never brought suddenly into 
 contact with any cold substance while they are hot. 
 
 78. Lamps. — The common spirit-lamp will be understood without 
 description from the figure (Fig. 13). When not actually lighted, the 
 wick must be kept covered with the glass cap ; for if 
 the wick were exposed to the air, the alcohol in the ^^^- ^^• 
 spirit upon it would evaporate faster than the water, 
 and the cotton would soon become water-soaked and 
 incapable of being lighted. 
 
 Whenever gas can be obtained, gas-lamps are greatly 
 to be preferred to the best spirit-lamps. For all ordi- 
 nary uses, the gas-lamp known as Bunsen's burner may 
 be employed. The cheapest and best construction of 
 the lamp may be learned from the following description 
 with the accompanying figure (Fig. 14). The single casting of brass 
 ab comprises the tube h through which the gas enters, and the block a 
 from which the gas escapes by two or three fine vertical holes passnig 
 through the screw d and issuing from the upper face of d, as shown at e. 
 The length of the tube 6 is 4.5 c. m., and its outside diameter varies 
 
174 
 
 BLAST-LAMPS. 
 
 [§79. 
 
 Fig. 14. 
 
 
 from 0.5 c. m, at the outer end to 1 c. m. at the junction with the block 
 a. The outside diameter of the block a is 1.6 c. m., and its outside 
 height without the screws is 1.8 c. m. By the screw c, the piece ab is 
 
 attached to the iron foot g, which may- 
 be 6 c. m. in diameter. By the screw d, 
 the brass tube / is attached to the cast- 
 ing ab. The diameter of the face e, 
 and therefore the internal diameter of 
 the tube /, should be 8 m. m. The 
 length of the tube / is 9 c. m. Through 
 the wall of this tube, four holes 5 m. m. 
 in diameter are to be cut at such a 
 height that the bottom of each hole 
 will come 1 m. m. above the face e 
 when the tube is screwed upon ab. 
 These holes are of course opposite each 
 other in pairs. The finished lamp is also shown in the figure. To 
 the tube b a caoutchouc tube of 5 to 7 m. m. internal diameter is 
 attached ; this flexible tube should be about 1 m. long, and its other 
 extremity should be connected with the gas- cock through the inter- 
 vention of a short piece of brass 
 gas-pipe screwed into the cock. In 
 cases where a very small flame is 
 required, as, for instance, in evapo- 
 rating small quantities of liquid, a 
 piece of wire-gauze somewhat larger 
 than the opening of the tube / should 
 be laid across the top of the tube, 
 and its projecting edges pressed 
 down tightly against the sides of the 
 tube before the gas is lighted. In 
 default of this precaution, the flame 
 of a Bunsen burner, when small and 
 exposed to currents of air, is liable 
 to pass down the tube and ignite the 
 gas at d. 
 
 79. Blast-lamps and Blowers. 
 
 — Though well suited for all the 
 
 ordinary operations of the laboratory, the lamps above described are 
 
 incapable of yielding a very intense heat. Hence, when the contents 
 
 of a platinum crucible are to be fused or intensely heated, a blast- 
 
 Fig. 15. 
 
79.] 
 
 BLOWERS. 
 
 175 
 
 lamp will be found useful. The best form is that sold under the name 
 
 of Bunsen's Gas Blowpipe. Its construction and the method of using 
 
 it may be learned from the accompanying figure (Fig. 15): ab is the 
 
 pipe through which the gas enters ; c is 
 
 the tube for the blast of air ; the relation ^^^- ^^^ 
 
 of the air tube to the external gas tube 
 
 is shown at d ; there is an outer sliding 
 
 tube by which the form and volume of 
 
 the flame can be regulated. 
 
 If gas is not to be had, a lamp burn- 
 ing oil or naphtha may be employed. Fig. 16 represents a glass- 
 blower's lamp, made of tin and suitable for burning oil. A large 
 wick is essential, whether oil or naphtha be the combustible. 
 
 For every blast-lamp a blowing- machine of some sort is necessary. 
 To supply a constant blast, it is essential that the bellows be of that 
 construction called double. 
 
 Figr. 17. 
 
 Fig. 17 represents a very good form of blowpipe-table. The bellows 
 are made of seamless rubber cloth ^ the table is 0,8 meter high, from 
 
17J5 
 
 BLOWERS. 
 
 [§79. 
 
 which the other dimensions may be inferred. A simpler form of 
 bellows, and one which can be made by any carpenter or cabinet- 
 maker, is represented in perspective and in section in Fig. 18. The 
 
 sides of the bellows and of the reservoir are made of stout leather. 
 The arrangement of valves will be evident from the figure ; a constant 
 pressure is maintained on the reservoir by means of a spiral spring, 
 and the air is delivered through the tube t. The rod which is repre- 
 sented in the figure serves simply as a guide. The entire length from 
 a to & may be 0.6 meter. 
 
 Where an abundant supply of water is at command, a Bunsen 
 pump of the kind described in App., § 76, but of larger size, may be 
 used to furnish a blast. The pipe above the 
 enlargement d (see Fig. 11) is left open to the 
 air instead of connecting with a flask as repre- 
 sented in Fig. 11. The waste-pipe k passes 
 through the cork of a large bottle, Fig. 19, of some 
 liters' capacity. Through the stopper of this 
 bottle there pass also two glass tubes ; one of 
 them, h, reaches nearly to the bottom of the 
 bottle and serves as a siphon ; the other merely 
 extends through the cork, and to it is attached 
 the tube ?', to convey. the blast to any desired 
 point. The water and air, which together flow 
 down the pipe kh, pass into the bottle. When 
 the water is turned on, the caoutchouc tube gi is 
 closed for a moment with the thumb and finger. 
 This starts the water through the siphon, and 
 immediately a continuous and po^verful blast of 
 air rushes out through the tube gi, and may be 
 carried directly to the blowpipe. The siphon 
 must be capable of carrying off a larger stream of water than that 
 which is allowed to enter, so that there shall never be more than 
 3 or 4 c. m. of water in the bottle. 
 
§79.] 
 
 BLOWERS. 
 
 177 
 
 Instead of the Bunsen Pump, the large-sized Richards' s Filter-Pump 
 may be used, if there is a sufficient head of water. The accompanying 
 figure (Fig. 20) shows the pump arranged to supply a blast, and the 
 
 Figr. 20. 
 
 WimUWUtK 
 
 L 
 
 dimensions of the blast attachment are given. The pump, with blast 
 attachment, can be obtained of dealers in laboratory supplies. 
 The efficiency of the blast depends upon the dimensions of the tubes 
 
178 
 
 LAMPS. 
 
 [§80. 
 
 and the head of water employed ; a fall of 6 or 8 feet furnishes an 
 efficient blast. 
 
 In default of a blast-lamp, platinum crucibles may readily be ignited 
 in a fire of coke or anthracite. To this end, place the tightly covered 
 platinum crucible in a somewhat larger crucible of refractory clay or 
 Hessian ware, and pack the space between the two crucibles tightly 
 with calcined magnesia, so that the platinum may nowhere come in 
 contact with the clay, ^over the coarse crucible, and place it, with its 
 contents, in the coal fire, in such a manner that it may be gradually 
 heated ; finally, imbed the crucible in the glowing coals and urge the 
 draught of the furnace for half an hour. The degree of heat to which 
 the contents of the platinum crucible may be exposed, in this way, in 
 an efficient fire, is really far greater than that of the blast-lamps above 
 described ; but the lamps are more convenient than the fire. 
 
 The effect of a simple Bunsen's burner may be greatly increased, 
 without the use of any blower, by surrounding its flame with a 
 cylinder of fire clay, 3 inches in diameter by 4 or 5 inches high, and 
 having walls at least f of an inch thick. The crucible, or other body 
 to be heated, is hung in the middle of this chimney, and is thus 
 exposed not only to the direct heat of the flame, but also to the radiant 
 heat from the clay walls which surround it. 
 
 Where no gas is to be had, an alcohol-lamp with circular wick, 
 
 of some one of the numerous 
 ^^^- 2^- forms sold under the name of 
 
 Berzelius' Argand Spirit-Lamp 
 (Fig. 21), will be found useful. 
 These argand lamps are usually 
 mounted on a lamp-stand pro- 
 vided with three brass rings ; 
 but the fittings of these lamps 
 are all made slender, in order 
 not to carry off too much heat. 
 When it is necessary to heat 
 heavy vessels, other supports 
 must be used. Several forms 
 of gasoline burners are at pres- 
 ent sold by dealers in laboratory 
 supplies. 
 
 80. Iron-stand, Tripod, 
 Wire-Gauze and Triangle. — To support vessels over the gas-lamp, 
 an iron-stand is used consisting of a stout vertical rod fastened into 
 
81.] 
 
 WATElt- AND SAND-BATB. 
 
 179 
 
 Fig. 22. 
 
 k« 
 
 Fig. 23. 
 
 a heavy cast-iron foot, and several iron rings of graduated sizes 
 secured to the vertical rod with binding screws ; all the rings may "be 
 slipped off the rod, or any ring may be adjusted at any 
 convenient elevation. The general arrangement is not 
 unlike that of the stand which makes part of the Ber- 
 zelius lamp (Fig. 21), although simpler and cheaper. 
 As a general rule, it is not best to apply the direct 
 flame to glass and porcelain vessels ; hence a piece of 
 iron wire-gauze of medium fineness is stretched loosely 
 over the largest ring, and bent downwards a little for 
 the reception of round-bottomed vessels ; on this gauze, 
 flasks and porcelain dishes are usually supported. Cru- 
 cibles, or dishes, too small for the smallest ring belong- 
 ing to the stand, are conveniently supported on an 
 equilateral triangle made of three pieces of soft iron 
 wire twisted together at the apices ; this triangle is 
 laid on one of the rings of the stand. An iron tripod — that is, a 
 stout ring supported on three legs — may often be used instead of the 
 stand above described, but it is not so generally 
 useful because of the difliculty of adjusting it at 
 various heights : with a sufficiency of wooden 
 blocks wherewith to raise the lamp or the tripod 
 as occasion may require, it may be made avail- 
 able. 
 
 81. Water-Bath and Sand-Bath. — It is 
 often necessary to evaporate solutions, or to dry 
 precipitates at a moderate temperature which 
 can permanently be kept below a certain known limit ; thus, when 
 an aqueous solution is to be quietly evaporated without spirting or 
 jumping, the temperature of the solution must 
 never be suffered to rise above the boiling-point of 
 water, nor even quite to reach this point. This quiet 
 evaporation is best effected by the use of a water- 
 bath, — a copper cup whose top is made of concentric 
 rings of different diameters to adapt it to dishes of 
 various sizes (Fig. 24). This cup, two-thirds full of 
 water, is supported on the iron-stand over the lamp, 
 and the dish containing the solution to be evaporated 
 is placed on that one of the several rings which will permit the greater 
 part of the dish to sink into the copper cup. The steam rising from 
 the water impinges upon the bottom of the dish, and brings the liquid 
 
 Fig. 24. 
 
180 
 
 BLOWPIPES, 
 
 t§82. 
 
 within it to a temperature which insures the evaporation of the water, 
 but will not cause any actual ebullition. The water in the copper cup 
 must never be allowed to boil away. Wherever a constant supply of 
 steam is at hand, as in buildings warmed by steam, the copper cup 
 above described may be converted into a steam-bath by attaching it 
 to a steam-pipe by means of a small tube provided with a stop-cock. 
 
 An empty tomato-can furnished with rings as above may take the 
 place of the copper cup; and, in fact, a cheap but serviceable water- 
 bath may be made from a quart milk-can, oil-can, tea-canister, or any 
 similarly shaped tin vessel, by inserting the stem of a glass funnel 
 into the neck of a can through a well-fitting cork. In this funnel the 
 dish containing the liquor to be evaporated rests. The can contains 
 the water, which is to be kept just boiling. On account of the shape 
 of the funnel, dishes of various sizes can be used with the same 
 apparatus. 
 
 When a gradual and equable heat higher than can be obtained upon 
 the water-bath is required, a sand-bath will sometimes be found useful. 
 A cheap and convenient sand-bath may be made by beating a disk of 
 thin sheet iron, about 4 inches in diameter, into the form of a 
 saucer or shallow pan, and placing within it a small quantity of dry 
 sand. The disk or flask to be heated is imbedded in the sand, and 
 the apparatus placed upon a ring of the iron-stand over a gas-lamp. 
 
 pj 25 ®2- Blowpipes. — The mouth-blowpipe in 
 
 its simplest form is a tube bent near one ex- 
 ^_/ri n tremity at a right angle. Fig. 25, a, represents 
 
 a common form of blowpipe used by jewellers. 
 The blowpipe is rendered more convenient by 
 the addition of a mouth-piece and a chamber 
 near the right angle for the condensation of 
 moisture. Fig. 25, b and c, represent different 
 forms of blowpipe thus furnished. The cheap- 
 est and best form of mouth-blowpipe for chemi- 
 cal purposes is a tube of tin-plate, about 18 
 c. m. long, 2 c. m. broad at one end, and tapering 
 to 0.7 c. m. at the other (Fig. 25, 6); the broad 
 end is closed, and serves to retain the moisture ; 
 a little above this closed end a small cylindrical 
 tube of brass about 5 c. m. long is soldered in at right angles ; this 
 brass tube is slightly conical at the end, and carries a small nozzle or 
 tip, which may be made either of brass or platinum. The tip should 
 be drilled out of a solid piece of metal, and should not be fastened 
 
 Qw*- 
 
 I 
 
 sss^Bbx 
 
 ^ 
 
82.] 
 
 BL O WPIPE-FLAME. 
 
 181 
 
 upon the brass tube with a screw. A trumpet-shaped mouth-piece 
 of horn or boxwood is a convenient, though by no means essential, 
 addition to this blowpipe. For convenience in cleaning and packing, 
 blowpipes are often made in several pieces, as is the one represented 
 in Fig. 25, c. 
 
 The blowpipe may be used with a candle, with gas or with any 
 hand-lamp proper for burning oil, petroleum or any of the so-called 
 burning fluids, provided that the form of the lamp below the wick- 
 holder is such as to permit the close approach of the object to be 
 heated to the side of the wick. When a lamp is used, a wick about 
 1.2 c. m. long and 0.5 c. m. broad is more convenient than a round or 
 narrow wick. The wick-holder should be filed off on its longer dimen- 
 sion a little obliquely, and the wick cut parallel to the holder, in order 
 that the blowpipe-flame may be directed downwards when necessary 
 (Figs. 26 and 27). A gas-flame suitable for the blowpipe is readily 
 obtained by slipping a narrow brass tube (i) open at both ends, into 
 the tube / of Bunsen's burner. (See Fig. 13.) This blowpipe-tube 
 must Ve long enough to close the air-apertures in the tube /, and 
 should be pinched together and filed off obliquely on top; it may 
 usually be obtained with the burner from dealers in chemical ware. 
 
 To use the mouth-blowpipe, place the open end of the tube between 
 the lips, or, if the pipe is provided with a mouth-piece, press the 
 trumpet-shaped mouth-piece against the lips ; fill the mouth with air 
 till the cheeks are widely distended, and insert the tip in the flame of 
 a lamp or candle ; close the communication between the lungs and 
 the mouth, and force a current of air through the tube by squeezing 
 the air in the mouth with the muscles of the cheeks, breathing, in the 
 meantime, regularly and quietly through the nostrils. The knack of 
 blowing a steady stream for several minutes at a time, is readily 
 acquired by a little practice. 
 
 It will be at once observed ^^S". 26. 
 
 that the appearance of the 
 flame varies considerably, 
 according to the strength of 
 the blast and the position of 
 the jet with reference to the 
 wick. 
 
 When the jet of the blow- 
 pipe is inserted into the mid- 
 dle of a candle-flame, or is placed in the lamp-flame in the position 
 shown in Fig. 26, and a strong blast is forced through the tube, a long, 
 
182 BLOWPIPE-FLAME. [§ 82. 
 
 blue cone of flame, a 6, is produced, beyond and outside of which 
 stretches a more or less colored outer cone towards c. The point of 
 greatest heat in this flame is at the point of the inner blue cone, 
 because the combustible gases are there supplied with just the quantity 
 of oxygen necessary to consume them, but between this point and the 
 extremity of the flame the combustion is concentrated and intense. 
 The greater part of the flame thus produced is oxidizing in its effect, 
 and this flame is technically called the oxidizing -flame. From the point 
 a of the inner blue cone, the heat of the flame diminishes in both direc- 
 tions, towards 6, on the one hand, and towards c on the other ; most 
 substances require the temperature which is found between a and c. 
 Oxidation takes place most rapidly at or just beyond the point c of 
 the flame, provided that the temperature at this point is high enough 
 for the special substance to be heated. 
 
 A flame of precisely the opposite chemical effect may be produced 
 with the blowpipe. To obtain a good redwcin^-flame, it is necessary 
 
 to place the tip of the blow- 
 Fig. 27. pipe, not within, but just 
 .^.7r^\ outside of the flame, and 
 "■^ to blow rather over than 
 
 through the middle of the 
 flame (Fig. 27). In this 
 manner, the flame is less 
 altered in its general char- 
 acter than in the former 
 case, the chief part con- 
 sisting of a large, luminous 
 cone, containing a quantity of free carbon in a state of intense igni- 
 tion, and just in the condition for taking up oxygen. This flame 
 is, therefore, reducing in its effect, and is technically called the 
 reducing-^duvue. The substance which is to be reduced by exposure 
 to this flame, should be completely covered up by the luminous 
 cone, so that contact with the air may be entirely avoided. It is 
 to be observed that, whereas to produce an effective oxidizing-flame 
 a strong blast of air is desirable, to get a good reducing-flame, the oper- 
 ator should blow gently, with only enough force to divert the lamp- 
 flame. 
 
 Substances to be heated in the blowpipe-flame are supported, some- 
 times on charcoal, and sometimes on platinum foil or wire, or in 
 platinum spoons or forceps. Charcoal is especially suitable for a 
 support in experiments in reduction. With reference to the choice 
 
§§83,84.] PLATINUM UTENSILS. l83 
 
 of charcoal for blowpipe experiments, see § 83. The manner of 
 holding the blowpipe is illustrated by Fig. 28. 
 
 83. Platinum Foil and Wire. Pincers. — A piece of platinum 
 foil about 1 ^ inches long, and 1 inch wide 
 
 will be suflBcient. The foil should be at ^^fif- 28. 
 
 least so thick that it does not crinkle 
 
 when wiped ; and it is more economical 
 
 to get foil which is too thick than too 
 
 thin, for it requires frequent cleaning. 
 
 To keep foil in good order it should be 
 
 frequently- scoured with fine moist sand, 
 
 and in case the foil becomes wrinkled, it 
 
 may be burnished by placing it upon the 
 
 bottom of an inverted agate or porcelain 
 
 mortar and rubbing it strongly with the 
 
 pestle. 
 
 A bit of platinum wire, not stouter than the wire of a small pin and 
 about 3 inches long, will last a long time with careful usage. It may 
 be cleaned by long-continued boiling in water. A small 
 loop, about as large as this O, should be bent at each end ^^S' 20. 
 of the wire. Q 
 
 When platinum foil is to be heated, it may be held at 
 one end with a pair of the small steel pincers known as 
 jewellers' tweezers. A piece of platinum wire, as long as 
 the one above •described, can be held in the fingers with- 
 out inconvenience, for platinum is, comparatively speaking, 
 a bad conductor of heat. Pieces of wire, too short to be 
 held, may be made serviceable by thrusting one end of 
 the wire into the end of a glass rod or closed tube which 
 has been softened in the blowpipe-flame. 
 
 84. Platinum Crucibles. — For several of the oper- 
 ations of quantit-ative analysis as now practised, platinum 
 crucibles are indispensable, and though not absolutely nec- 
 essary for the profitable study of qualitative analysis, one Q 
 of these vessels will often be found convenient by the stu- 
 dent of the elements of analysis. It will be well, therefore, for the 
 student, who proposes to continue his chemical studies beyond quali- 
 tative analysis, to procure a platinum crucible once for all. A crucible 
 of the capacity of about 20 cubic centimeters will be large enough for 
 most uses ; it should be cylindrical rather than flaring, and should be 
 
184 PLATINUM UTENSILS. [§§86,86. 
 
 provided with a loose cover in the form of a shallow dish. For the 
 purposes of this book, however, a small crucible holding 7 or 8 centi- 
 meters answers every purpose and a cover is not essential. 
 
 No other metal, and no mixture of substances from which a metal 
 can be reduced, must ever be heated in a platinum crucible, or upon 
 platinum foil or wire, for platinum forms alloys with other metals, and 
 these alloys are much more fusible than platinum itself. If once 
 alloyed with a baser metal, the platinum ceases to be applicable to its 
 peculiar uses. 
 
 Platinum may be cleansed by boiling it in either nitric or hydrochloric 
 acid, by fusing acid sulphate of sodium upon it, or by scouring it with 
 fine sand. Aqua regia and chlorine- water dissolve platinum; the 
 sulphides, cyanides, and hydrates of sodium and potassium, when 
 fused in platinum vessels, slowly attack the metal. 
 
 85. Wash-bottle. — A wash-bottle is a flask with a uniformly thin 
 bottom closed with a sound cork or caoutchouc stopper through which 
 pass two glass tubes, as shown in Fig. 30. 
 
 The outer end of the longer tube is drawn to a 
 Fig. 30. moderately fine point. A short bend near the bottom 
 of this longer tube in the same plane and direction as 
 the upper bend is of some use, because it enables the 
 operator to empty the flask more completely by inclin- 
 ing it. By blowing into the short tube, a stream of 
 water will be driven out of the long tube with consid- 
 erable force. This force with which the stream is pro- 
 jected adapts the apparatus to removing precipitates 
 from the sides of vessels as well as to washing them 
 on filters. For use in analytical operations, it is often 
 convenient to attach a caoutchouc tube 12 or 15 c. m. 
 long to the tube through which the air is blown ; this 
 flexible tube should be provided with a glass mouth- 
 piece, consisting of a bit of glass tubing about 3 c. m. long. As the 
 wash-bottle is often filled with hot or even boiling water, it may be 
 improved by binding about its neck a ring of cork, or winding the 
 neck closely with smooth cord. It may then be handled without 
 inconvenience when hot. 
 
 The method of making a wash-bottle is described in the following 
 paragraphs. 
 
 86. Glass Tubing. — Two qualities of glass tubing are used in 
 chemical experiments, that which softens readily in the flame of a gas- 
 
§§ 87, 88.] CUTTING AND CRACKING GLASS. 185 
 
 or spirit-lamp, and that which fuses with extreme difficulty in the flame 
 of the blast-lamp. These two qualities are distinguished by the terms 
 soft and hard glass. Soft glass is to be preferred for all uses except 
 
 87654 3 2 1 
 
 the intense heating, or ignition, of dry substances. Fig. 31 represents 
 the common sizes of glass tubing, both hard and soft, and shows also 
 the proper thickness of the glass walls for each size. The numbers 
 ranging from 4 to 8 are best suited for use in qualitative analysis. 
 
 87. Stirring- rods. — Cut a short stick of glass rod. No. 8 or 7, into 
 pieces four or five inches long (see the next paragraph), and round 
 the sharp ends by fusion in the blowpipe-flame. 
 
 88. Cutting and Cracking Glass. — Glass tubing and glass rod 
 must generally be cut to the length required for any particular appa- 
 ratus. A sharp triangular file is used for this purpose. The stick of 
 tubing, or rod, to be cut is laid upon a table, and a deep scratch is 
 made with the file at the place where the fracture is to be made. The 
 stick is then grasped with the two hands, one on each side of the 
 mark, while the thumbs are brought together just at the scratch. By 
 pushing with the thumbs and pulling in the opposite direction with 
 the fingers, the stick is broken squarely at the scratch, just as a stick 
 of candy or dry twig may be broken. The sharp edges of the frac- 
 ture should invariably be made smooth, either with a wet file, or by 
 softening the end of the tube or rod in the lamp (App., § 89). Tubes 
 or rods of sizes 4 to 8 inclusive may readily be 
 
 cut in this manner ; the larger sizes are divided with Figr. 32. 
 
 more difficulty, and it is often necessary to make 
 the file- mark both long and deep. An even fracture 
 is not always to be obtained with large tubes. The 
 lower ends of glass funnels, and those ends of gas 
 delivery-tubes which enter the bottle or flask in 
 which the gas is generated, should be filed off, or 
 ground off on a grindstone, obliquely (Fig. 32), to facilitate the drop- 
 ping of liquids from such extremities. 
 
 rig*. o;«s. 
 
186 MANIPULATION OF GLASS, [§ 89. 
 
 In order to cut glass plates, the glazier's diamond must be resorted 
 to. For the cutting of exceedingly thin glass tubes and of other glass- 
 ware, like flasks, retorts and bottles, still other means are resorted to, 
 based upon the sudden and unequal application of heat. The process 
 divides itself into two parts, the producing of a crack in the required 
 place, and the subsequent guiding of this crack in the desired direc- 
 tion. To produce a crack, a scratch must be made with the file, and 
 to this scratch a pointed bit of red-hot charcoal, or the jet of flame 
 produced by the mouth-blowpipe, or a very fine gas-flame, or a red- 
 hot glass rod may be applied. If the heat does not produce a crack, a 
 wet stick or file may be touched upon the hot spot. Upon any part of 
 a glass surface except the edge, it is not possible to control perfectly the 
 direction and extent of this first crack ; at an edge a small crack may be 
 started with tolerable certainty by carrying the file-mark entirely over 
 the edge. To guide the crack thus started, a pointed bit of charcoal 
 or slow-match may be used. The hot point must be kept on the glass 
 from 1 c. m. to 0.5 c. m. in advance of the point of the crack. The 
 crack will follow the hot point, and may therefore be carried in any 
 desired direction. By turning and blowing upon the coal or slow- 
 match, the point may be kept sufficiently hot. Whenever the place 
 of experiment is supplied with common illuminating gas, a very small 
 jet of burning gas may be advantageously substituted for the hot coal 
 or slow-match. To obtain such a sharp jet, a piece of hard glass tube, 
 No. 4, 10 c, m. long, and drawn to a very fine point (App., § 89), should 
 be placed in the caoutchouc tube which ordinarily delivers the gas to 
 the gas-lamp, and the gas should be lighted at the fine extremity. 
 The burning jet should have a fine point, and should not exceed 
 1.5 c. m. in length. By a judicious use of these simple tools, broken 
 tubes, beakers, flasks, retorts and bottles may often be made to yield 
 very useful articles of apparatus. No sharp edges should be allowed 
 to remain upon glass apparatus. The durability of the apparatus 
 itself, and of the corks and caoutchouc stoppers and tubing used with 
 it, will be much greater, if all sharp edges are removed with the file, 
 or, still better, rounded in the lamp. 
 
 89. Bending and Closing Glass Tubes. — Tubing of sizes 4 
 to 8 inclusive can generally be worked in the common gas- or 
 spirit-lamp; for larger tubes the blast-lamp is necessary (App., § 79). 
 Glass tubing must not be introduced suddenly into the hottest part 
 of the flame, lest it crack. Neither should a hot tube be taken from 
 the flame and laid at once upon a cold surface. Gradual heating and 
 gradual cooling are alike necessary, and are the more essential the 
 
§89.] MANIPULATION OF GLASS, 187 
 
 thicker the glass ; very thin glass will sometimes bear the most 
 sudden changes of temperature, but thick glass and glass of uneven 
 thickness absolutely require slow heating and annealing. When the 
 end of a tube is to be heated, as in rounding sharp edges, more 
 care is required in consequence of the great facility with which cracks 
 start at an edge. A tube should, therefore, always be brought first 
 into the current of hot air beyond the actual flame of the gas- or spirit- 
 lamp, and there thoroughly warmed, before it is introduced into the 
 flame itself. If a blast-lamp is employed, the tube may be warmed in 
 the smoky flame before the blast is turned on, and may subsequently 
 be annealed in the same manner ; the deposited soot will be burnt off 
 in the first instance, and in the last may be wiped off when the tube 
 is cold. In heating a tube, whether for bending, drawing or closing, 
 the tube must be constantly turned between the fingers, and also moved 
 a little to the right and left, in order that it may be uniformly heated 
 all round, and that the temperature of the neighboring parts may be 
 duly raised. If a tube or rod is to be heated at any part but an end, 
 it should be held between the thumb and first two fingers of each 
 hand in such a manner that the hands shall be below the tube or rod, 
 with the palms upward, while the lamp-flame is between the hands. 
 When the end of a tube or rod is to be heated, it is best to begin by 
 warming the tube or rod about 2 c. m. from the end, and thence to 
 proceed slowly to the end. 
 
 The best glass will not be blackened or discolored during heating. 
 Blackening occurs in glass which, like ordinary flint glass, contains 
 lead as an ingredient. Glass containing much lead is not well adapted 
 for chemical uses. The blackening may sometimes be removed by 
 putting the glass in the upper or outer part of the flame, where the 
 reducing- gases are consumed, and the air has the best access to 
 the glass. The blackening may be altogether avoided by always 
 keeping the glass in the oxidizing part of the flame. 
 
 Glass begins to soften and bend below a visible red heat. The 
 condition of the glass is judged of as much by the fingers as the eye ; 
 the hands feel the yielding of the glass, either to bending, pushing or 
 pulling, better than the eye can see the change of color or form. It 
 may be bent as -soon as it yields in the hands, but can be drawn out 
 only when much hotter than this. Glass tubing, however, should not 
 be bent at too low a temperature ; the curves made at too low a heat 
 are apt to be flattened, of unequal thickness on the convex and concave 
 sides, and brittle. 
 
 In bending tubing to make gas delivery- tubes and the like, attention 
 
188 
 
 MANIPULATION OF GLASS. 
 
 [§89. 
 
 Pig. 33. 
 
 should be paid to the following points : 1st, the glass should be equally 
 hot on all sides ; 2d, it should not be twisted, pulled out or pushed 
 together during the heating ; 3d, the bore of the tube at the bend 
 should be kept round, and not altered in size ; 4th, if two or more 
 bends be made in the same piece of tubing 
 (Fig. 33, a), they should all be in the same 
 /"r^ (C a l l pl^-ne, so that the finished tube will lie flat 
 
 i> I upon the level table. 
 
 U When a tube or rod is to be bent or drawn 
 
 J f close to its extremity, a temporary handle 
 
 may be attached to it by softening the end 
 of the tube or rod, and pressing against the soft glass a fragment of 
 glass tube, which will adhere strongly to the softened end. The handle 
 may subsequently be removed by a slight blow, or by the aid of a file. 
 If a considerable bend is to be made, so that the angle between the 
 arms will be very small or nothing, as in a siphon, for example, the 
 curvature cannot be well produced at one place in the tube, but should 
 be made by heating, progressively, several centimeters of the tube, and 
 bending continuously from one end of the heated portion to the other 
 (Fig. 33, 6). Small and thick tube maybe bent more sharply than 
 large or thin tube. 
 
 A lamp for bending glass tubing, better than the ordinary form of 
 the Bunsen burner, is one the tube of which is flattened out so as to 
 give a thin but broad flame of the same character as the ordinary 
 lamp, but in shape more like a bat-wing burner. 
 (See Fig. 34.) The tube is placed in this flame 
 and turned round and round until it reaches the 
 proper temperature ; it is then withdrawn from 
 the flame and bent. In this way a regular curve 
 may be obtained and the sides of the tube do not 
 collapse. 
 
 In order to draw a glass tube down to a finer 
 bore, it is simply necessary to thoroughly soften on 
 all sides one or two centimeters' length of the tube, 
 and then taking the glass from the flame, pull the 
 parts asunder by a cautious movement of the hands. 
 The larger the heated portion of glass, the longer 
 will be the tube thus formed. Its length and fineness also increase 
 with the rapidity of motion of the hands. If it is desirable that the 
 finer tube should have thicker walls in proportion to its bore than 
 the original tube, it is only necessary to keep the heated portion soft 
 
 Pig. 34. 
 
§90.] MANIPULATION OF GLASS. 189 
 
 for two or three minutes before drawing out the tube, pressing the 
 parts slightly together the while. By this process the glass will be 
 thickened at the hot ring. 
 
 To obtain a tube closed at one end, it is best to take a piece of 
 tubing, open at both ends, and long enough 
 to make two closed tubes. In the middle of 
 the tube a ring of glass, as narrow as pos- 
 sible, must be made thoroughly soft. The 
 hands are then separated a little, to cause 
 a contraction in diameter at the hot and 
 soft part. The point of the flame must now 
 
 be directed, not upon the narrowest part of the tube, but upon what 
 is to be the bottom of the closed tube. This point is indicated by the 
 line a in Fig. 35. By withdrawing the right hand, the narrow part 
 of the tube is attenuated, and finally melted off, leaving both halves 
 of the original tube closed at one end, but not of the same form ; 
 the right-hand half is drawn out into a long point, the other is more 
 roundly closed. It is not possible to close handsomely the two pieces 
 at once. The tube is seldom perfectly finished by the operations ; a 
 superfluous knob of glass generally remains upon the end. If small 
 it may be got rid of by heating the whole end of the tube, and blowing 
 moderately with the mouth into the open end. The knob, being hot- 
 ter, and therefore softer than any other part, yields to the pressure 
 from within, spreads out and disappears. If the knob is large, it may 
 be drawn off by sticking to it a fragment of tube, and then softening 
 the glass above the junction. The same process may be applied to 
 the too pointed end of the right-hand half of the original tube, or 
 to any misshapen result of an unsuccessful attempt to close a tube, 
 or to any bit of tube which is too short to make two closed tubes. 
 When the closed end of a tube is too thin, it may be strengthened by 
 keeping the whole end at a red heat for two or three minutes, turning 
 the tube constantly between the fingers. It may be said in general 
 of all the preceding operations before the lamp, that success depends 
 on keeping the tube to be heated in constant rotation, in order to 
 secure a uniform temperature on all sides of the tube. 
 
 90. Blowing Bulbs. — Bulb-tubes, like the one represented in 
 Fig. 36, are employed for reducing substances capable of forming 
 sublimates upon the cold walls of the tube. They are readily made 
 from bits of tubing, in the flame of Bunscn's burner, or in the common 
 blowpipe-flame. 
 
 If the bulb desired is large in proportion to the size of the tube on 
 
190 • CAOUTCHOUC. [§91. 
 
 which it is to be made, the walls of the tube must be thickened by 
 rotation in the flame before the bulb can be blown. The thickened 
 portion of glass is then to be heated to a cherry -red, suddenly with- 
 drawn from the flame, and expanded 
 Fig. 36. while hot by steadily blowing, or 
 
 rather pressing, air into the tube 
 with the mouth ; the tube must be 
 constantly turned on its axis, not 
 only while in the flame, but also 
 while the bulb is being blown. If 
 too strong or too sudden a pressure 
 be exerted with the mouth, the bulb will be extremely thin and quite 
 useless. By watching the expanding glass, the proper moment for 
 arresting the pressure may usually be determined. If the bulb 
 obtained be not large enough, it may be reheated and enlarged by 
 blowing into it again, provided that a sufficient thickness of glass 
 remain. If a bulb is to be blown in the middle of a piece of tubing, 
 the thickening is effected by gently pressing the ends of the tube to- 
 gether while the glass is red-hot in the place where the bulb is to be. 
 
 It is sometimes necessary to make a hole in the side of a tube or 
 other thin glass apparatus. This may be done by directing a pointed 
 flame from the blast- lamp upon the place where the hole is to be, 
 until a small spot is red-hot, and then blowing forcibly into one end 
 of the tube while the other end is closed by the finger ; at the hot 
 spot the glass is blown out into a thin bubble, which bursts or may 
 be easily broken off, leaving an aperture in the side of the tube. 
 
 It is hoped that these few directions will enable the attentive student 
 to perform, sufficiently well, all the manipulations with glass tubes 
 which the experiments described in this manual require. Much practice 
 will alone give a perfect mastery of the details of glass-blowing. 
 
 91. Caoutchouc. — Vulcanized caoutchouc is a most useful sub- 
 stance in the laboratory, on account of its elasticity and because it 
 resists so well most of the corrosive substances with which the chemist 
 deals. It is used in three forms: (1) in tubing of various diameters 
 comparable with the sizes of glass tubing ; (2) in stoppers of various 
 sizes to replace corks ; (3) in sheets. Caoutchouc tubing may be 
 used to conduct all gases and liquids which do not corrode its sub- 
 stance, provided that the pressure under which the gas or liquid flows 
 be not greater, or their temperature higher, than the texture of the 
 tubing can endure. The flexibility of the tubing is a very obvious 
 advantage in a great variety of cases. Short pieces of such tubing, a 
 
§92.] CAOUTCHOUC. — CORKS. 191 
 
 few centimeters in length, are much used, under the name of con- 
 nectors, to make flexible joints in apparatus, of which glass tubing 
 forms part; flexible joints add greatly to the durability of such 
 apparatus, because long glass tubes bent at several angles and con- 
 nected with heavy objects, like globes, bottles or flasks full of liquid, 
 are almost certain to break even with the most careful usage; gas 
 delivery-tubes, and all considerable lengths of glass tubing, should 
 invariably be divided at one or more places, and the pieces joined 
 again with caoutchouc connectors. The ends of the glass tubing to 
 be thus connected should be squarely cut, and then rounded in the 
 lamp, in order that no sharp edges may cut the caoutchouc ; the inter- 
 nal diameter of the caoutchouc tube must be a little smaller than the 
 external diameter of the glass tubes ; the slipping on of the connector 
 is facihtated by wetting the glass. In some cases of delicate quanti- 
 tative manipulations, in which the tightest possible joints must be 
 secured, the caoutchouc connector is bound on to the glass tube with 
 a silk or smooth linen string ; the string is passed as tightly as possible 
 twice round the connector and tied with a square knot ; the string 
 should be moistened in order to prevent it from slipping while the 
 knot is tying. 
 
 Caoutchouc stoppers are much more durable than corks, and are in 
 every respect to be preferred, if of proper shape and good quality. 
 Caoutchouc stoppers can be bored, like corks (see the next section), 
 by means of suitable cutters, and glass tubes can be fitted into the 
 holes thus made with a tightness unattainable with corks ; but these 
 stoppers may be bought already provided with one, two and three 
 holes. It is not well to lay in a large stock of caoutchouc stoppers, for 
 though they last a long time when in constant use, they not infre- 
 quently deteriorate when kept in store, becoming hard and somewhat 
 brittle with age. 
 
 Pieces of thin sheet caoutchouc are very conveniently used for 
 making tight joints between large tubes of two different sizes, or 
 between the neck of a flask, or bottle, and a large tube which enters 
 it, or between the neck of a retort and the receiver into which it 
 enters. A sufficiently broad and long piece of sheet caoutchouc is 
 considerably stretched, wrapped tightly over the glass parts adjoining 
 the aperture to be closed, and secured in place by a string wound 
 closely about it and tied with a square knot. 
 
 92. Corks. — It is often very difficult to obtain sound, elastic corks 
 of fine grain and of size suitable for large flasks and wide-mouthed 
 bottles. On this account, bottles with mouths not too large to be 
 
192 
 
 CORKS. 
 
 [§92. 
 
 closed with a cork cut across the grain, should be chosen for chemical 
 uses, in preference to bottles which require large corks or bungs cut 
 with the grain, and therefore offering continuous channels for the 
 passage of gases, or even liquids. The kinds sold as champagne corks 
 and as satin corks for phials are suitable for chemical use. The best 
 corks generally need to be softened before using ; this softening may 
 be effected by rolling the cork under a board upon the table, or under 
 the foot upon the clean floor, or by gently squeezing it on all sides 
 with the well-known tool expressly adapted for this purpose, and thence 
 called a cork-squeezer. Steaming also softens the hardest corks. 
 
 Corks must often be cut with cleanness and precision ; a sharp, 
 thin knife, such as shoemakers use, is desirable for this purpose. 
 When a cork has been pared down to reduce its diameter, a flat file 
 may be employed in finishing ; the file must be fine enough to leave 
 a smooth surface upon the cork ; in filing a cork, a cylindrical, not a 
 conical, form should be aimed at. 
 
 In boring holes through corks to receive glass tubes, a hollow 
 cylinder of sheet brass sharpened at one end is a very convenient 
 tool. Fig. 37 represents a set of such little cylinders of graduated 
 sizes, slipping one within the other into a very 
 compact form ; a stout wire, of the same length 
 3 as the cylinders, accompanies the set, and 
 serves a double purpose, — passed transversely 
 through two holes in the cap which terminates 
 ^^. each cylinder, it gives the hand a better grasp 
 
 [ I of the tool while penetrating the cork ; and 
 
 I I when the hole is made, the wire thrust through 
 
 ^-r-r an opening in the top of the cap expels the 
 
 U little cylinder of cork which else would remain 
 
 in the cutting cylinder of brass. That cutter, 
 whose diameter is next below that of the glass 
 tube to be inserted in the cork, is always to be 
 selected, and if the hole it makes is too small, 
 a round file must be used to enlarge the aper- 
 ture ; the round file, also, often comes in play 
 to smooth the rough sides of a hole made by a dull cork-borer. A 
 pair of small calipers is a very convenient, though by no means es- 
 sential, tool in determining which size of cutter to employ. A flask 
 which presents sharp or rough edges at the mouth can seldom be 
 tightly corked, for the cork cannot be introduced into the neck without 
 being cut or roughened; such sharp edges must be rounded in the 
 
 Pier. 37. 
 
§93.] 
 
 GA S-GENERA TOES. 
 
 193 
 
 lamp. In thrusting glass tubes through bored corks, the following 
 directions are to be observed : (1) The end of the tube must not pre- 
 sent a sharp edge capable of cutting the cork. (2) The tube should be 
 grasped very close to the cork, in order to escape cutting the hand 
 which holds the cork, should the tube break ; by observing this pre- 
 caution, the chief cause of breakage, viz. irregular lateral pressure, 
 will be at the same time avoided. (3) A funnel-tube must never be 
 held by the funnel in driving it through a cork, nor a bent tube grasped 
 at the bend, unless the bend comes immediately above the cork. 
 (4) If the tube goes very hard through the cork, the application of a 
 little soap and water will facilitate its passage, but if soap is used, the 
 tube can seldom be withdrawn from the cork after the latter has 
 become diy. (5) The tube must not be pushed straight into the cork, 
 but screwed in, as it were, with a slow rotary as well as onward 
 motion. Joints made with corks should always be tested before the 
 apparatus is used, by blowing into the apparatus, and at the same 
 time stopping up all legitimate outlets. Any leakage is revealed by 
 the disappearance of the pressure created. To the same end, air may 
 be sucked out of an apparatus and its tightness proved by the per- 
 manence of the partial vacuum. To attempt to use a leaky cork is 
 generally to waste time and labor, and to insure the failure of the 
 experiment. 
 
 93. Gas-bottle. — Fig. 38 represents a gas-bottle fitted for evolv- 
 ing sulphuretted hydrogen, carbonic acid and other gases which can 
 be prepared without heat. A straight glass tube of 
 convenient length is slipped into the caoutchouc 
 connector at the right to carry the gas into the 
 solution to be tested. The neck of the bottle 
 should be rather narrow, since it is difficult to 
 obtain tight stoppers for bottles with wide mouths, 
 but must nevertheless be wide enough to admit a 
 cork, or better a caoutchouc stopper, capable of 
 carrying both the delivery and the thistle-tubes. 
 
 To prepare, for example, sulphuretted hydrogen 
 gas, put a tablespoonful of fragments of sulphide of 
 iron in the bottom of the bottle, replace the cork 
 with its tubes, and press, or rather twist, it tightly 
 into the neck of the bottle ; pour in enough water 
 through the thistle-tube to seal the lower end of 
 that tube, and finally as much concentrated sulphuric acid as would 
 be equal to a tenth or a twelfth of the volume of the water. 
 
194 
 
 GAS-GENEBATORS. 
 
 [§94. 
 
 At the start it is well thus to mix strong acid with the water in the 
 bottle, for the heat generated by the union of the two liquids serves to 
 warm the apparatus, and to facilitate the decomposition of the sul- 
 phide of iron ; but it must be remembered that strong sulphuric acid 
 is by itself unfit for generating sulphuretted hydrogen, and that the 
 evolution of gas would be checked if much of it were added. When 
 the flow of gas ceases, pour a small portion of dilute sulphuric acid 
 into the thistle-tube, and repeat this operation as often as may be 
 necessary to maintain a constant stream of gas. Dilute acid fit for 
 this purpose may be prepared by mixing 1 volume of strong sulphuric 
 acid with 14 volumes of water ; — the water should be well stirred and 
 the acid poured into it in a fine stream. 
 
 In precipitating the members of Classes II and III with sulphu- 
 retted hydrogen, the gas delivery-tube should not dip deeper than 
 about an inch beneath the surface of the liquid in the beaker. A 
 rapid current of gas is useless and wasteful. The best method of 
 operating is to pour dilute sulphuric acid into the thistle- tube in such 
 quantity that the bubbles of gas may follow one another slowly enough 
 to be counted without effort. 
 
 94. Self- regulating Gas- generator. — An apparatus which is 
 always ready to deliver a constant stream of sulphuretted hydrogen, 
 and yet does not generate the gas except 
 when it is immediately wanted for use, 
 is a great convenience in an active labo- 
 ratory. The same remark applies to the 
 two gases, hydrogen and carbonic acid, 
 which are likewise used in considerable 
 quantities in quantitative analysis, and 
 which can be conveniently generated in 
 precisely the same form of apparatus 
 which is advantageous for sulphuretted 
 hydrogen. Such a generator may be 
 made of divers dimensions. The fol- 
 lowing directions, with the accompanying 
 figure (Fig. 39), will enable the student 
 to construct an apparatus of convenient 
 size. Procure a glass cylinder 20 or 25 
 c, m, in diameter and 30 or 35 c. m. high ; 
 ribbed candy jars are sometimes to be had of about this size ; pro- 
 cure also a stout tubulated bell-glass 10 or 12 c. m. wide and 5 or 7 c. m. 
 shorter than the cylinder. Get a basket of sheet-lead 7.6 c. m. deep 
 
 Fig. 39. 
 
§ 94.] GAS-GEN EBATOBS. 195 
 
 and 2.5 c. m. narrower than the bell-glass, and bore a number of small 
 holes in the sides and bottom of this basket. Cast a circular plate of 
 lead 7 m. m. thick and of a diameter 4 c. m. larger than that of the 
 glass cylinder; on what is intended for its under side solder three 
 equidistant leaden strips, or a continuous ring of lead, to keep the 
 plate in proper position as a cover for the cylinder. Fit tightly to 
 each end of a good brass gas-cock a piece of brass tube 8 c. m. long, 
 1.5 to 2 cm. wide, and stout in metal. Perforate the centre of the 
 leaden plate, so that one of these tubes will snugly pass through the 
 orifice, and secure it by solder, leaving 5 c. m. of the tube projecting 
 below the plate. Attach to the lower end of this tube a stout hook on 
 which to hang the leaden basket. By means of a sound cork and 
 common sealing-wax, or a cement made of oil mixed with red and 
 white lead, fasten this tube into the tubulure of the bell-glass air-tight, 
 and so firmly that the joint will bear a weight of several pounds. 
 Hang the basket by means of copper wire within the bell 5 c. m. above 
 the bottom of the latter. To the tube which extends above the ' stop- 
 cock attach by a good cork the neck of a tubulated receiver of 100 or 
 150 c. c. capacity, the interior of which has been loosely stuffed with 
 cotton. Into the second tubulure of the receiver fit tightly the delivery- 
 tube carrying a caoutchouc connector; into this connector can be 
 fitted a tube adapted to convey the gas in any desired direction. 
 When many persons use the same generator, each person must bring 
 his own tube. 
 
 To charge the apparatus, fill the cylinder with dilute acid to within 
 10 or 12 c. m. of the top, fill the basket with fragments of sulphide of 
 iron, hang the basket in the bell, and put the bell-glass full of air into 
 its place with the stop- cock closed. On opening the cock, the weight 
 of the acid expels the air from the bell, the acid comes in contact with 
 the solid in the basket, and a steady supply of gas is generated un I 
 either the acid is saturated or the solid dissolved ; if the cock 1 . ) 
 closed, the gas accumulates in the bell, and pushes the acid below t? ) 
 basket so that all action ceases. In cold weather the apparatus mu *". 
 be kept in a warm place. For generating sulphuretted hydroger, 
 sulphuric acid diluted with 14 parts of water is used ; for hydro- 
 gen, zinc and sulphuric acid diluted with 4 or 5 parts of water; 
 while for carbonic acid, marble and hydrochloric acid diluted with 2 
 or 3 parts of water should be taken. 
 
 Many forms of self -regulating gas- generators can be obtained of 
 dealers in chemical apparatus : that known as " Kipp's Generator " is 
 excellent for use in small laboratories. 
 
196 MOBTABS. [§95. 
 
 95. Mortars. — Whenever the substance to be analyzed occurs in 
 the form of large pieces or coarse powder, it should, as a general rule, 
 be pulverized by mechanical means before subjecting it to the action 
 of solvents. Mortars of iron, steel, agate or porcelain are used for 
 this purpose, according to the character of the substance to be 
 powdered. 
 
 An iron mortar is useful for coarse work and for effecting the first 
 rough breaking up of substances which are subsequently powdered in 
 the agate or porcelain mortar. If there be any risk of fragments 
 being thrown out of the mortar, it should be covered with a cloth or 
 piece of stiff paper, having a hole in the middle through which the 
 pestle may be passed. Instead of the common iron mortar a small 
 steel mortar, of the kind called diamond mortars by dealers in chemical 
 ware, may be used for crushing minerals. Pieces of stone, minerals 
 and lumps of brittle metals may be safely broken into fragments 
 suitable for the mortar by wrapping them in strong paper, laying them 
 so enclosed upon an anvil, and striking them with a heavy hammer 
 The paper envelope retains the broken particles which might otherwise 
 fly about in a dangerous manner, and be lost. 
 
 The best porcelain mortars are those known by the name of Wedge- 
 wood- ware, but there are many cheaper substitutes. Porcelain mortars 
 will not bear sharp and heavy blows; they are intended rather for 
 grinding or triturating saline substances than for hammering ; the 
 pestle may either be formed of one piece of porcelain, or a piece of 
 porcelain cemented to a wooden handle ; the latter is the less desira- 
 ble form of pestle. Unglazed porcelain mortars are to be preferred. 
 In selecting mortars, the following points should be attended to : 
 1st, the mortar should not be porous ; it ought not to absorb strong 
 acids or any colored fluid, even if such liquids be allowed to stand for 
 hours in the mortar ; 2d, it should be very hard, and its pestle should 
 be of the same hardness ; 3d, it should be sound ; 4th, it should have 
 a lip for the convenience of pouring out liquids and fine powders. As 
 a rule, porcelain mortars will not endure sudden changes of tempera- 
 ture. Tliey may be cleaned by rubbing in them a little sand soaked in 
 nitric or sulphuric acid, or if acids are not appropriate, in caustic soda. 
 
 Agate mortars are only intended for trituration ; a blow would 
 break them. They are exceedingly hard, and impermeable. The 
 material is so precious and so hard to work, that agate mortars are 
 always small. The pestles are generally inconveniently short, — a 
 difficulty which may be remedied by fitting the agate pestle into a 
 wooden handle. 
 
§96.] SPATULA. 197 
 
 In all grinding operations in mortars, whether of porcelain or agate, 
 it is expedient to put only a small quantity of the substance to be 
 powdered into the mortar at once. The operation of powdering will 
 be facilitated by sifting the matter as fast as it is powdered, returning 
 to the mortar the particles which are too large to pass through the 
 sieve. 
 
 96. Spatulae. — For transferring substances in powder, or in small 
 grains or crystals, from one vessel to another, spatulae and scoops 
 made of horn or bone are convenient tools. A coarse bone paper- 
 knife makes a good spatula for laboratory use. Cards, free from 
 glaze and enamel, are excellent substitutes for spatulae. 
 
INDEX. 
 
 Acetate of lead, as reagent, 156. 
 
 sodium, as reagent, 154. 
 Acetates, tests for, 103. 
 Acetic acid, as reagent, 150. 
 Acid solutions, 122, 124. 
 Alcohol, as reagent, 158. 
 Alkaline solutions, 122. 
 Alloys, treatment of, 141. 
 Aluminates, precipitated with Class 
 
 IV, 46. 
 Aluminum, confirmatory test for, 50. 
 a member of Class IV, 
 
 12,46. 
 precipitated as hydrate, 
 
 12,46. 
 presence of, indicated, 56. 
 Ammonia-water, as reagent, 151. 
 Ammonium salts, testing for, 109. 
 
 indicated, 109. 
 Antimony, confirmatory tests for, 37. 
 converted into an insolu- 
 ble oxide by HNO3, 143. 
 see sulphide of. 
 globule brittle, 112. 
 a member of Class III, 
 
 11, 32. 
 precipitated as sulphide, 
 
 11, 32. 
 presence of, indicated, 42, 
 
 125. 
 spots distinguished from 
 
 arsenic spots, 37. 
 trioxide of, as sublimate, 
 109. 
 Aqua regia, how prepared, 150. 
 
 its use as a solvent, 120. 
 Argand spirit-lamp, 178. 
 Arseniates, test for, 92. 
 
 Arsenic, confirmatory test for, 35. 
 
 a member of Class III, 11, 
 
 32. 
 presence of, indicated, 30, 
 
 42, 113. 
 see sulphide of. 
 separation a& sulphide, 11, 
 
 32. 
 as a sublimate, 110. 
 Arsenic acid, tests for, 35, 92. 
 Arsenious acid, distinguished from 
 
 arsenic acid, 92. 
 Arsenious oxide as sublimate, 110. 
 Arsenites, tests for, 92. 
 Ash of charcoal, 112. 
 
 Barium, a member of Class VI, 15. 
 precipitated as carbonate, 
 
 05. 
 precipitated as chromate, 
 
 66. 
 salts soluble in ammoniacal 
 
 solutions, 81. 
 test for certain classes of 
 salts, 79. 
 Beakers, 165. 
 
 Biborate of sodium (borax), as re- 
 agent, 154. 
 Bichromate of potassium, as reagent, 
 
 154. 
 Bismuth, confirmatory test for, 26. 
 globule brittle, 112. 
 a member of Class II, 11, 23. 
 precipitated as hydrate, 26. 
 precipitated as sulphide, 
 
 11, 23. 
 presence of, indicated, 112, 
 125. 
 
 199 
 
200 
 
 INDEX. 
 
 Bisulphide of carbon, as reagent, 158. 
 Blast-lamps, 174. 
 Blowing bulbs, 189. 
 Blowpipe, 180. 
 
 how to use, 181. 
 lamp, 181. 
 Boracic acid, precipitated from an al- 
 kaline solution, 123. 
 tests for, 96. 
 Borates, tests for, %. 
 Borax-bead, how to dilute, 133. 
 how to make, 133. 
 Bottles, how to handle, 163. 
 Bromides, tests for, 98, 100. 
 Bromine, tests for, 98, 100. 
 Bunsen's burner, 173. 
 
 filter-pump, 172. 
 
 Cadmium, a member of Class II, 11. 
 precipitated as sulphide, 
 
 11, 28. 
 presence of, indicated, 30, 
 
 125. 
 separation of, 28. 
 Calcium, a member of Class VI, 14. 
 precipitated as carbonate, 
 
 14,65. 
 precipitated as oxalate, 67. 
 test for certain classes of 
 salts, 82. 
 Caoutchouc stoppers, 190. 
 
 tubing, 191. 
 Carbonate of ammonium, as reagent, 
 
 152. 
 Carbonate of sodium, as reagent, 
 
 how purified, 153. 
 Carbonates, tests for, 88. * 
 Carbonic acid, tests for, 88. 
 Charcoal for blowpipe use. 111. 
 Chlorates, tests for, 103. 
 Chloride of ammonium, as reagent, 
 152. 
 of ammonium, uses of, 13, 
 56, 69. 
 Chloride of barium, as reagent, 156. 
 Chloride of calcium, as reagent, how 
 
 prepared, 155. 
 Chloride of lead, solubility of, 11. 
 Chloride of mercury, as reagent, 
 157. 
 
 Chloride of silver soluble in ammo- 
 nia-water, 21. 
 Chlorides, tests for, 84, 98. 
 Chlorine, tests for, 98. 
 Chlorine water, as reagent, 158. 
 Chromate of lead, a test for chro- 
 mium, 49. 
 Chromate of potassium, as reagent, 
 
 154. 
 Chromates, barium test for, 80. 
 
 reduction of, by HjS, 31, 
 
 91. 
 tests for, 91. 
 Chrome iron ore, hard to decompose. 
 
 133. 
 Chromic oxide, hard to decompose, 
 
 133. 
 Chromites, precipitated with Class 
 
 IV, 46. 
 Chromium, a member of Class IV, 
 12, 46. 
 detected as chromate of 
 
 sodium, 49. 
 gives a green borax 
 
 bead, 133. 
 precipitated as hydrate, 
 
 12,46. 
 presence of, indicated, 
 56. 
 Class, the term defined, 0. 
 Class I, defined, 6. 
 
 how to precipitate, 19. 
 Class II, defined, 7. 
 
 how to precipitate, 23. 
 Class III, defined, 7. 
 
 precipitation of, 32. 
 Class IV, defined, 12. 
 
 how to precipitate, 47, 55. 
 salts precipitated with, 46. 
 Class V, defined, 13. 
 
 how to precipitate, 58, 63. 
 Class VI, defined, 14. 
 
 how to precipitate, 65, 69. 
 Class VII, defined, 15. 
 
 how isolated, 15, 73. 
 Clay chimney for Bunsen's burner, 
 
 178. 
 Closed-tube test, 106. 
 Cobalt, member of Class V, 14, 58. 
 blowpipe test for, 61. 
 
INDEX. 
 
 201 
 
 Cobalt, confirmatory test for, 62. 
 
 precipitated as sulphide, 13, 
 
 58. 
 presence of, indicated, 63, 
 115. 
 Copper, member of Class II, 11, 23. 
 confirmatory test for, 27. 
 gives blue solutions, 26. 
 globule described, 112, 113. 
 precipitated as sulphide, 11, 
 
 23. 
 presence of, indicated, 31. 
 Cork-cutters, 192. 
 Corks, 191. 
 
 to force tubes through, 193. 
 Cyanide of mercury, to detect cyano- 
 gen in, 89. 
 Cyanide of potassium, reagent, 154. 
 Cyanides, tests for, 88, 89. 
 Cyanogen, tests for, 88, 89. 
 
 Deflagration, 139. 
 Dissolving in acids, 118. 
 in water, 117. 
 
 Effervescence, what it indicates, 
 
 87. 
 Elements, identified by compounds, 2. 
 
 treated of, 1. 
 Etching glass, a test for fluorine, 97. 
 Evaporation test applied to a liquid, 
 
 145. 
 
 Ferri- and Fbrro-cyanide of po- 
 tassium, as reagents, 154. 
 Ferric chloride, as reagent, 156. 
 Filtering, 165. 
 Filter-stand, 167. 
 Filtration, rapid, 167. 
 Flasks, 164. 
 Fluoride of silicon, a test for fluorine 
 
 and silicon, 97. 
 Fluorides, tests for, 96. 
 Folding filters, 16(5. 
 Funnels, 165. 
 
 Fused minerals, how treated, 135. 
 Fusion with acid sulphate of sodium, 
 139. 
 
 with CaCOs and NH4CI, 139. 
 
 with carbonate of sodium, 134. 
 
 Fusions in platinum crucibles, 114. 
 
 Gas bottle, 193. 
 
 Gas-generators, self-regulating, 194. 
 General reagent, defined, 7. 
 General reagents for acids, 78. 
 
 to be used in a cer- 
 tain order, 16. 
 Glass, bending and closing tubes, 186. 
 
 cutting and cracking, 185. 
 
 tubing, 184. 
 Gold, globule described, 112. 
 
 insoluble in HNO3, 142. 
 
 a member of Class III, 11, 39. 
 
 presence of, indicated, 43. 
 
 confirmatory test for, 41. 
 
 purple of Cassius, test for, 143. 
 
 Hydrates, presence of, indicated, 
 
 104. 
 Hydrochloric acid, as reagent, 149. 
 its application as 
 a solvent, 118. 
 Hydrogen, its presence inferred, 76. 
 Hyposulphites (thiosulphates), tests 
 for, 90. 
 
 Indigo solution, how prepared, 157. 
 Insoluble substances, 130. 
 Iodides, tests for, 99, 100. 
 Iodine, tests for, 99, 100. 
 lodo-starch paper, 155. 
 Iron, discrimination between ferrous 
 and ferric salts, 54. 
 
 to be converted into ferric salt, 
 54. 
 
 a member of Class IV, 12, 46. 
 
 precipitated as hydrate, 12, 46. 
 
 presence of, indicated, 56. 
 
 Prussian blue, test for, 51. 
 
 reaction in borax bead, 134. 
 Iron stand, 178. 
 
 Labelling, importance of, 9. 
 
 Lamps, 173. 
 
 Lead, belongs to two classes, 11. 
 
 globule described, 112. 
 
 a member of Class I, 7, 19. 
 
 a member of Class II, 11, 23. 
 
 paper, how prepared, 156. 
 
202 
 
 INDEX. 
 
 Lead, precipitated as sulphide, 11, 
 23. 
 precipitation as chloride, 19. 
 precipitation as sulphate, 20, 
 
 26. 
 presence of, indicated, 30. 
 Lime-water, as reagent, 155. 
 Liquids to be tested with litmus, 
 
 145. 
 Litmus paper, how prepared, 157. 
 
 Magnesium, as member of Class 
 VII, 15, 71. 
 precipitated as phos- 
 phate of Mg and 
 NH4, 15, 71. 
 separation as oxide, 73. 
 mixture, as reagent, 
 156. 
 Manganate of sodium, 51. 
 Manganese, conhrmatory test, 51. 
 
 precipitated with Class 
 
 IV, 47. 
 precipitation of, as hy- 
 drate, 59. 
 presence indicated, 63. 
 Mercuric chloride, as reagent, 157. 
 Mercurous chloride, reaction with 
 
 ammonia-water, 21. 
 Mercury, belongs to two classes, 11. 
 compounds as sublimates, 
 
 109. 
 member of Class I, 7, 19. 
 member of Class II, 11, 23. 
 precipitation as subchlo- 
 
 ride, 19. 
 presence of, indicated, 30. 
 reduction of the metal in a 
 
 closed tube, 21. 
 reduction of the metal on 
 
 copper, 25. 
 separation as sulphide, 11, 
 
 23. 
 as a sublimate, 109. 
 Metallic elements, seven classes of, 17. 
 globules described, 112. 
 globules tested in oxidizing 
 flame, 113. 
 Metals, action of HNO3 on, 141. 
 used in the arts, 141. 
 
 Molybdate of ammonium, a test for 
 phosphates, 93. 
 of ammonium, as reagent, 
 how prepared, 152. 
 
 Mortars, 196. 
 
 Neutral solutions, 122. 
 Nickel, blowpipe test for, 61. 
 
 member of Class V, 14, 58. 
 precipitated as cyanide, 61. 
 precipitated as hydrate, 61. 
 precipitated as sulphide, 14, 
 
 58. 
 presence of, indicated, 63, 
 115. 
 Nitrate of barium, as reagent, 156. 
 when used, 82. 
 of cobalt, as reagent, 156. 
 of potassium, as reagent, 155. 
 of silver, as reagent, 155. 
 of sodium, as reagent, 154. 
 Nitrates, tests for, 102. 
 Nitric acid, action on metals, 121. 
 dilute, as reagent, 150. 
 strong, as reagent, 150. 
 tests for, 102. 
 when to be used as a 
 solvent, 118, 119, 141. 
 Nitrite of potassium, as reagent, how 
 
 prepared, 155. 
 Nitrogen peroxide, 108. 
 Non-metallic elements, how detected, 
 76. 
 
 Organic matter, detection of, 107. 
 how destroyed, 
 115. 
 Organic substances hinder the pre- 
 cipitation of Class IV, 57. 
 Oxalate of ammonium, reagent, 152. 
 Oxalates converted into carbonates, 
 68. 
 tests for 95. 
 Oxalic acid as a sublimate, 109. 
 as reagent, 150. 
 tests for, 95. 
 
 oxides, recognition of, 
 104. 
 Oxide of manganese, as reagent, 156. 
 Oxide of mercury, as reagent, 157. 
 Oxidizing blowpipe-flame, 182. 
 
INDEX. 
 
 203 
 
 Oxygen, how recognized, 108. 
 
 its presence inferred, 76. 
 
 Peroxides, how recognized, 104. 
 Phosphate of calcium, presence of, 
 indicated, 50. 
 of sodium, as reagent, 154. 
 Phosphates, precipitated with Class 
 IV, 46. 
 tests for, 93. 
 Phosphoric acid, tests for, 93. 
 Pincers, 183. 
 Platinic chloride, as reagent, how 
 
 prepared, 157. 
 Platinum, a member of Class III, 11, 
 32. 
 crucibles, 183. 
 foil, 183. 
 
 insoluble in HNO3, 142. 
 presence of, indicated, 43. 
 test for, 40, 143. 
 wire, 183. 
 Porcelain crucibles, 172. 
 
 dishes, 172. 
 Potassium, precipitated as chloro- 
 platinate, 73. 
 flame-test for, 72. 
 a member of Class VII, 
 15, 71. 
 Precipitates compacted by boiling 
 
 and shaking, 58. 
 Preliminary examination of a liquid, 
 
 99. 
 Preliminary treatment, 106. 
 
 of a pure metal or alloy, 
 141. 
 Prussian blue, a test for iron, 51. 
 Pulverizing, 196. 
 
 Qualitative analysis defined, 1. 
 
 Reagent bottles, 162. 
 Reagents, 149-158. 
 Reducing blowpipe-flame, 182. 
 Reduction test. 111. 
 
 how to perform with 
 
 delicacy, 114. 
 to be applied to in- 
 soluble substances, 
 131. 
 
 Richards's aspirator, 172. 
 
 Salts, kinds of, considered, 78. 
 
 soluble in water, 120. 
 Sand-bath, 179. 
 Separation of two elements, 5. 
 Silicates, alkaline, decomposed by 
 acids, 96, 123. 
 decomposed by chloride 
 of ammonium, 139. 
 Silicates, tests for, 96. 
 
 the commonest of insoluble 
 substances, 138. 
 Silicic acid, precipitated from an al- 
 kaline solution, 123. 
 Silver, a member of Class I, 6, 19. 
 globule described, 112. 
 precipitated by reducing 
 
 agents, 86. 
 precipitation as chloride, 19. 
 salts, insoluble in dilute nitric 
 
 acid, 85. 
 salts, sundry, colors of, 85. 
 see chloride of. 
 test for certain classes of 
 salts, 83. 
 Slaked lime, as reagent, 155. 
 Sodium, crystallization of chloro- 
 platinate of, 73. 
 flame-test for, 72. 
 a member of Class VII, 15, 
 
 71. 
 hydrate as reagent, 152. 
 Solubilities, table of, 128, 129. 
 Solution of indigo, as reagent, 157. 
 Solutions of known composition, 
 
 159-161. 
 Solvents, the order of use, 117. 
 Spatulae, 197. 
 
 Special tests for non-metallic ele- 
 ments, 87-104. 
 Stannous chloride, as reagent, how 
 
 prepared, 156. 
 Starch paste, how prepared, 158. 
 Starch-test for bromine, 98. 
 
 iodine, 99. 
 Stirring-rods, 185. 
 Stoppers, stuck, how to loosen, 163. 
 Strontium, a member of Class VI, 15, 
 65. 
 
204 
 
 INDEX. 
 
 Strontium, precipitated as carbonate, 
 15,66. 
 precipitated as sulpliate, 
 67. 
 Sublimates on charcoal while reduc- 
 ing metals, 113. 
 Sulphate of copper, as reagent, 156. 
 of sodium, acid, how pre- 
 pared, 154. 
 Sulphates, barium test for, 81. 
 how to detect, 93. 
 insoluble, reduced to sul- 
 phides, 132. 
 Sulphate of potassium, as reagent, 
 
 how prepared, 154. 
 Sulphide of ammonium, as reagent, 
 
 how prepared, 151. 
 Sulphide of arsenic, oxidation of, 34, 
 
 41. 
 Sulphide of sodium, as reagent, how 
 
 prepared, 153. 
 Sulphide of tin, oxidation of, 34. 
 Sulphides, tests for, 89. 
 Sulphides of arsenic, as sublimates, 
 
 110. 
 Sulphites, tests for, 90. 
 Sulphur, as a sublimate, 110. 
 Sulphuretted hydrogen, decomposed 
 
 by oxidizing agents, 30. 
 Sulphuretted hydrogen, how pre- 
 pared, 151. 
 Sulphuretted hydrogen, how to em- 
 ploy it, 8, 23. 
 Sulphuretted hydrogen, necessity of 
 
 expelling, 54. 
 Sulphuretted hydrogen water, how 
 
 prepared and kept, 151. 
 Sulphuric acid, as reagent, 150. 
 
 tests for, 93. 
 Sulphurous acid, 90. 
 Sulphur, precipitation of, 22, 31. 
 
 Table for Class I, 22. 
 
 Table for Class II, 29. 
 
 III, 39. 
 
 IV, 53. 
 
 V, 62. 
 
 VI, 68. 
 
 the separation of the seven 
 classes of metallic ele- 
 ments, 75. 
 Table of solubilities, 128, 129. 
 
 the seven classes of metallic 
 elements, 17. 
 Tartaric acid, as reagent, 151. 
 
 tests for, 75. 
 Tartrates, tests for, 95. 
 Test-tube rack, 164. 
 Test-tubes, 163. 
 Thiosulphates, tests for, 90. 
 Tin, a member of Class III, 11, 32. 
 confirmatory test for, 38. 
 converted into an insoluble ox- 
 ide by HNO3, 34. 
 globule, described, 112. 
 presence of, indicated, 42, 43. 
 reduction of, by zinc, 38. 
 see sulphide of. 
 test for, 38. 
 Triangle, 178. 
 Tripod, 178. 
 
 Utensils, list of, 162. 
 
 Water, 158. 
 Water-bath, 179. 
 Wash-bottle, 184. 
 Wire-gauze, 178. 
 
 Zinc, a member of Class V, 13, 58. 
 as reagent, 157. 
 precipitated as sulphide, 13, 
 
 58, 60. 
 presence of, indicated, 63. 
 confirmatory test for, 60. 
 
QDSS Eliot, C.W. 48515 
 E42n The compendious manual of 
 1892 qualitative chemical anal- 
 ysis^ 16th ed. 
 

 
 >.:'iy f '