LIBRARY 
 
 OF THE 
 
 UNIVERSITY OF CALIFORNIA. 
 
 Deceived ^-^^s .189 2_ 
 
 tAccexsions No<^^^. Cla-ss No. 
 
THE 
 
 ORGANIC ANALYSIS OF POTABLE 
 WATERS 
 
THE ORGANIC ANALYSIS 
 
 OF 
 
 POTABLE WATEES 
 
 BY 
 
 J. A. BLAIR, M.B., C.M., D.SC.EDIN. 
 L.R.C.P.LoND. 
 
 LONDON 
 J. & A. CHURCHILL 
 
 11, NEW BURLINGTON STREET 
 
 1890 
 
$ 
 
CONTENTS 
 
 PAGE 
 
 PREFACE . . . . . . vii 
 
 INTRODUCTION . . . . .1 
 
 PART I. The Albuminoid Ammonia Process, and Oxygen Pro- 
 cess at Ordinary Temperatures . . 5- 
 PABT II. The Oxygen Process at 100 C. . . .31 
 
 PART III. The Sulphuric Acid Process for Organic Nitrogen . 55 
 PART IV. The Sulphuric Acid and Permanganate Process for 
 
 Organic Carbon . . . .83 
 
 APPENDIX A. Solutions for Albuminoid Ammonia Process . 103 
 B. Solutions for Oxygen Process at Ordinary 
 
 Temperatures .... 106 
 
 C. Solutions for Oxygen Process at 100 C. . 109 
 
 -, D. Estimation of Nitrates by Arsenious Acid . 110 
 
 ,, E. Solutions for the Sulphuric Acid and Perman- 
 ganate Process for Organic Carbon . . 117 
 
PREFACE 
 
 THE greater part of the experimental work contained 
 in the following pages was done in the Laboratory of 
 the Medical Jurisprudence and Public Health Depart- 
 ment of the University of Edinburgh. I gratefully 
 acknowledge the kindly interest shown, and en- 
 couragement given by Sir Douglas Maclagan whilst 
 I was engaged with the work. 
 
 It was Dr Hunter Stewart, of the Public Health 
 Department, who suggested to me the possibility 
 of proving that Kjeldahl's Process for estimating 
 Organic Nitrogen can be applied to potable waters; 
 and he also pointed out to me the possible utility of 
 a series of analyses of potable waters made by the 
 "Oxygen and Albuminoid Ammonia Processes." I 
 gratefully acknowledge not only these suggestions of 
 his, but also the information given by him, from time 
 
viii Preface, 
 
 to time, on subjects bearing on the work in general, 
 and the references given by him to the literature of 
 the subject in this, and other countries. 
 
 J. A. BLAIR. 
 
 MAIDSTONB ; 
 
 July, 1890. 
 
THE 
 
 ORGANIC ANALYSIS OF POTABLE 
 WATERS 
 
 INTRODUCTION 
 
 THIS branch of analytical work is handicapped by 
 two great drawbacks, namely, the very small quantity 
 of organic matter which has to be dealt with, and the 
 impossibility, in the present state of chemical science, 
 of measuring these organic bodies as such. Much of 
 the organic matter in potable waters is highly com- 
 plex in composition, unstable, and readily induced 
 to undergo chemical changes, if the conditions of its 
 existence are modified. In order to meet these diffi- 
 culties, analysts have resorted to analytical methods, 
 whereby these highly complex bodies are broken up 
 into more elementary, and more stable compounds. 
 Nitrogen is a constituent of many organic compounds 
 more especially of albuminoid bodies, and, there- 
 fore, has been used as a partial indicator of the 
 
 1 
 
The Organic Analysis 
 
 quantity of organic matter contained in a sample of 
 water. Carbon is also present in many organic 
 bodies, and is also used as a partial indicator of the 
 quantity of organic matter. Speaking generally, 
 organic bodies whilst decomposing do not take up 
 either nitrogen or carbon, therefore analysts prefer to 
 measure the quantity of these elements, rather than 
 the quantity of hydrogen and oxygen, which also are 
 contained in very many organic compounds. Many 
 organic compounds undergoing decomposition absorb 
 oxygen, and the quantity of oxygen thus absorbed 
 may be used to measure the relative quantity of 
 organic matter present in the water examined. 
 
 The very small amount of organic matter, in a 
 reasonable quantity of average potable waters, has 
 forced chemists to resort to very delicate processes of 
 analysis. This delicacy of operation tends to make 
 the average experimental error considerable, even in 
 trained hands. The experimental error is least with 
 the analytical process which assures accuracy of 
 measurement, consistent with simplicity of apparatus 
 and the minimum of precautions. 
 
 In this country, the organic impurity of potable 
 waters is usually estimated by one of three methods, 
 or by a combination of them, namely : 
 
 (1) The Albuminoid Ammonia Process brought 
 out by Messrs. Wanklyn, Chapman, and Smith (1867). 
 
 (2) The Estimation of Organic Carbon and Nitro- 
 
of Potable Waters 3 
 
 gen brought out by Drs Frankland and Armstrong 
 (1868). 
 
 (3) The Oxygen Process as elaborated by Prof. 
 Tidy (1879). 
 
 The various analytical processes will be considered 
 in the following order : 
 
 I. The Albuminoid Ammonia and Oxygen Pro- 
 cesses, and then some considerations in reference to 
 the analyses of waters by both methods. 
 
 II. The Oxygen Process at 100 C. 
 
 III. The Estimation of the Organic Nitrogen by 
 the Sulphuric Acid Process. 
 
 IV. The Estimation of the Organic Carbon by the 
 Sulphuric Acid and Permanganate Process. 
 
PART I 
 
 THE ESTIMATION OF THE ORGANIC IMPUEITY OF 
 POTABLE WATEES BY THE "ALBUMINOID AM- 
 MONIA PROCESS," AND BY THE " OXYGEN PRO- 
 CESS AT ORDINARY TEMPERATURES " 
 
 THESE two processes are similar in only giving 
 partial results. The " Oxygen Process," as usually 
 carried out, does not completely oxidise all the organic 
 matter; and the " Albuminoid Ammonia Process" does 
 not convert all the organic nitrogen, which may be 
 present in a water, into ammonia. These partial 
 measurements of oxygen required, and nitrogen pre- 
 sent are in some respects unsatisfactory. Putting 
 aside the question of micro-organisms and specific 
 poisons, sanitarians seem to agree that the danger to 
 health, from drinking a water containing organic 
 matter, does not depend so much upon the quantity 
 of organic matter contained in the water, as upon the 
 capability of the matter for readily undergoing chemi- 
 cal changes. The albuminoid ammonia and oxygen 
 
6 The Organic Analysis 
 
 processes, to some extent, measure the potentiality of 
 the organic matter for rapid chemical changes. 
 
 Both methods take advantage of the fact that 
 organic matter, in the presence of suitable oxidising 
 agents, readily takes up oxygen, and becomes partially 
 oxidised into less complex, but more chemically stable, 
 compounds. In the oxygen process, the quantity of 
 oxygen used up in this way is made the indicator for 
 judging of the organic impurity of the sample of 
 water; but in Mr Wanklyii's process, the ammonia 
 evolved by the combined action of caustic alkali and 
 an oxidising agent is the indicator of the amount of 
 organic impurity. It is claimed for both processes 
 that the results obtained are sufficiently uniform to 
 enable the analyst to form an opinion of the fitness, or 
 unfitness of the water for potable purposes. 
 
 The Albuminoid Ammonia Process may be carried 
 out in the following way. 
 
 Certain solutions are required, and instructions for 
 their preparation will be found in Appendix A. 
 
 The solutions are 
 
 (1) A dilute solution of ammonium chloride of the 
 strength that each c.c. contains -j-J-^- milligramme of 
 ammonia. 
 
 (2) Nessler's reagent. 
 
 (3) A strongly alkaline solution of permanganate 
 ,of potash. 
 
 (4) A saturated solution of sodium carbonate. 
 
of Potable Waters 
 
 Take a 32 oz. distillation flask, with a side tube in 
 the neck, and cleanse it well with good tap water ; 
 if not previously used for this purpose, wash out the 
 flask with strong sulphuric acid, and again with tap 
 water till the drippings are neutral to litmus paper. 
 Fit the flask with a good, well-fitting cork covered 
 with a layer of tin-foil, and attach the distillation 
 flask to a clean Liebig condenser, in the usual way, 
 packing with tin-foil or writing-paper, if necessary, 
 and supporting the flask on wire-gauze over the ring 
 of an iron stand. It is 'very necessary to have all 
 measuring flasks and apparatus scrupulously clean. 
 Measure out half a litre of the water to be analysed; 
 pour it into the flask through a long funnel. If 
 the water is acid, add 10 c.c. of the sodium car- 
 bonate solution. Boil the water briskly, and collect 
 the distillate in four stoppered flasks of 50 c.c. ca- 
 pacity. The first 50 c.c. of the distillate is now poured 
 into a nesslerising glass, and 2 c.c. of Nessler's reagent 
 added to it ; mix well by gently shaking the glass, 
 and, if ammonia is present, it will strike a yellow colour. 
 Into another nesslerising glass run in from a burette 
 as much of the weak solution of ammonium chloride 
 as you think equal to the ammonia in the distillate, 
 make up to 50 c.c. with distilled water free of 
 ammonia- and add 2 c.c. of Nessler's reagent. If 
 sufficient ammonium chloride has not been added to 
 give the same colour as in the 50 c.c. of distillate, run 
 
8 The Organic Analysis 
 
 in rapidly successive quantities of ammonium chloride 
 solution till the colours correspond ; when the proper 
 quantity of solution required has been thus found, 
 repeat the operation in a clean nesslerising glass. 
 When nesslerising, stand the glasses on a white tile. 
 The "Free Ammonia" is estimated in this way, and Mr 
 Wanklyn recommends adding to the total quantity 
 one third of the quantity found, and in this way 
 avoiding the necessity of nesslerising the remaining 
 150 c.c. of distillate. 
 
 The Albuminoid Ammonia. Take 50 c.c. of the 
 alkaline permanganate solution, boil it in a small 
 flask for five minutes, in order to free it of traces of 
 ammonia, and then add it to the 300 c.c. of the water 
 left in the distillation flask. Resume the boiling pro- 
 cess, and collect the distillate in successive quantities 
 of 50 c.c. ; nesslerise each of these 50 c.c. in the 
 above manner till the last 50 c.c. contain no ammonia, 
 then stop any further distillation. 
 
 The free and albuminoid ammonia found in a half- 
 litre of water, if doubled will give the ammonia as 
 parts per million, or milligrammes per litre. 
 
of Potable Waters 
 
 The Oxygen Process at Ordinary Temperatures. 
 
 Solutions required (for preparation see Appendix 
 B): 
 
 (1) Solution of potassium permanganate 10 c.c. 
 of the solution are capable of yielding up one milli- 
 gramme of oxygen. 
 
 (2) Dilute sulphuric acid. 
 
 (3) Solution of potassium iodide. 
 
 (4) Solution of sodium hyposulphite. 
 
 (5) Solution of starch. 
 
 Clean out with strong sulphuric acid four flasks of 
 about 20 oz. capacity, and then rinse out with good 
 tap water. Into each of two of these flasks pour 
 250 c.c. of the water to be examined, and into the 
 other two flasks pour 250 c.c. of distilled water; to 
 each flask add 10 c.c. of the dilute sulphuric acid and 
 10 c.c. of the permanganate solution ; cork the flasks 
 with clean corks or cover with paper, and place them 
 at a window exposed to a good light. Note the time 
 when the permanganate is added to the waters, and 
 if the pink colour of the water become very pale 
 before the allotted time is up, it is advisable to add 
 another 10 c.c. of permanganate solution. At the 
 end of a quarter of an hour, remove one flask con- 
 taining the sample of water and one containing dis- 
 tilled water, and add to each flask five drops of the 
 
10 The Organic Analysis 
 
 solution of potassium iodide for each 10 c.c. of per- 
 manganate added. Eotate the flasks gently till the 
 colour is a clear, brilliant yellow. From a burette 
 run into the flask containing distilled water the hypo- 
 sulphite solution till the yellow colour is very faint ; 
 now add sufficient starch solution to produce a deep 
 blue colour; continue to run in slowly the hypo- 
 sulphite solution till the blue colour just disappears, 
 and note the quantity of hyposulphite solution used. 
 Treat the sample of water in exactly the same way. 
 At the end of three hours, treat the contents of the 
 remaining two flasks in the same way. 
 
 Calculation required. In all cases, the quantity of 
 hyposulphite needed to decolorise the iodine is pro- 
 portionate to the quantity of permanganate in the 
 water, or, in other words, to the quantity of oxygen 
 available for oxidising purposes. In the blank expe- 
 riment with distilled water, the quantity of hypo- 
 sulphite used represents one milligramme of oxygen, 
 or 10 c.c. of permanganate solution. If the number 
 of c.c. of hyposulphite solution used by the sample of 
 water is deducted from the number of c.c. of hypo- 
 sulphite solution used by the distilled water, the 
 difference will represent the oxygen used up by the 
 sample of water. 
 
 The equation is stated thus : c.c. of hyposul- 
 phite used by distilled water : c.c. of difference : : 
 one milligramme of oxygen I answer . , % : 
 
of Potable Waters 11 
 
 _ c.c. of difference _ oxygen used expressed in the 
 c.c. of blank exp. ~ fraction of a milligramme. 
 
 The chemical equations, representing the reactions 
 which take place at the various stages of the process, 
 are given in Appendix B. 
 
 Sulphuretted hydrogen, ferrous salts, and nitrites 
 will reduce permanganate. In the case of waters 
 containing any one of these chemical compounds, in 
 order to prevent errors arising in the estimation of 
 the oxygen used by the organic matter, Prof. Tidy 
 recommends heating the acidified water to get rid of 
 the sulphuretted hydrogen, and the estimation of the 
 oxygen used up in five minutes when ferrous salts, 
 or nitrites are present. The oxygen thus used is 
 deducted from the amount of oxygen used up in a 
 quarter of an hour and in three hours, in order to find 
 the quantity used up by the organic matter. Divide 
 the oxygen used by 2' 5, and the answer gives the 
 oxygen as parts per 100,000 of the water. 
 
 The following tables give a series of analyses of 
 various organic solutions and waters by the albu- 
 minoid ammonia and oxygen processes. 
 
12 
 
 The Organic Analysis 
 
 TABLE I 
 Animal Organic Solutions. 
 
 Solution I of fresh White of Egg, made by dis- 
 solving 2-518 grammes of the fresh material, and I'lO 
 grammes of freshly ignited sodium carbonate in 200 
 c.c. of distilled water. Solution was kept in a stop- 
 pered flask. 
 
 Solution II of fresh White of Egg, made by dis- 
 solving 0-525 gramme of the fresh material, and 0'50 
 gramme of sodium carbonate in 100 c.c. of distilled 
 water. The solution was kept in a stoppered flask. 
 
 Solution III of fresh White of Egg, made by dfs- 
 solving 1-388 grammes of the fresh material, and 1'39 
 grammes of freshly ignited sodium carbonate in 100 
 c.c. of distilled water. 
 
 Ammonia and oxygen are stated as parts per 
 million. 
 
 No. of 
 solution. 
 
 Age of 
 solution. 
 
 Quantity 
 used. 
 
 Oxygen 
 hour. 
 
 used in 
 
 Ammonia 
 yielded. 
 
 3 hours. 
 
 I 
 
 6 days 
 
 07 c.c. 
 
 100-0 
 
 178-8 
 
 158-5 
 
 11 
 
 3 
 
 4-0 
 
 33-9 
 
 47.2 
 
 49-0 
 
 II 
 
 5 
 
 3-0 
 
 34-4 
 
 48-4 
 
 48-0 
 
 II 
 
 5 
 
 2-0 
 
 34-6 
 
 47-3 
 
 42-0 
 
 III 
 
 Iday 
 
 i-o 
 
 77-6 
 
 103-2 
 
 180-0 
 
of Potable Waters 
 
 13 
 
 TABLE II 
 Water polluted with Organic Animal Matter. 
 
 TO c.c. of Fresh Urine was added to 2 litres of 
 Edinburgh City water. 
 
 Oxygen and albuminoid ammonia are stated as 
 milligrammes per litre, or parts per million. 
 
 Water. 
 
 Age. 
 
 Quantity used. 
 
 Oxygen used in 
 
 Alb. 
 ammonia 
 yielded. 
 
 Tidy. 
 
 Wanklyn. 
 
 | hour. 3 hours. 
 
 Edinburgh City 
 
 Fresh 
 
 250 c.c. 
 
 500 c.c. 
 
 0-82 
 
 2-15 
 
 0-17 
 
 
 1 day old 
 
 50 
 
 20 
 
 1-29 
 
 2-95 
 
 5-20 
 
 
 2 days old 
 3 
 5 
 
 7 
 
 50 
 50 
 50 
 100 
 
 20 
 20 
 20 
 
 50 
 
 1-13 
 
 1-09 
 0-98 
 1-16 
 
 2-57 
 2-42 
 2-51 
 
 2-49 
 
 3-55 
 
 3-47 
 
 3-08 
 
 
 12 
 
 100 
 
 50 
 
 0-94 
 
 2-52 
 
 3-96 
 
 1 part above "1 
 
 
 
 
 
 
 
 sample plus 2 > 
 
 14 
 
 200 
 
 100 
 
 0-38 
 
 0-72 
 
 0-68 
 
 parts aq. dest. J 
 
 
 
 
 
 
 
 Sewage made by making a 1 per cent, solution of 
 fresh urine in distilled water. 
 
 Oxygen and albuminoid ammonia stated as parts 
 per million. 
 
 Water. 
 
 Age. 
 
 Quantity 
 used. 
 
 Oxygen 
 
 f ' 
 | hour. 
 
 used in 
 3 hours. 
 
 Alb. 
 ammonia 
 yielded. 
 
 Sewage 
 
 Fresh 
 
 1-0 c.c. 
 
 13-30 
 
 24-30 
 
 90-00 
 
14 
 
 The Organic Analysis 
 
 TABLE III 
 Effluent Waters. 
 
 No. 1. Sewage water taken from the main sewer 
 as it enters Craigentinny Meadows. Heavy rains a 
 few hours previously. 
 
 No. 2. Effluent water from Graigentinny Meadows, 
 just before it runs into the sea. 
 
 No. 3. Edinburgh City water polluted with the 
 above effluent water (No. 2). 
 
 Oxygen and albuminoid ammonia are stat^i as 
 parts per million, or milligrammes per litre. 
 
 No. and age. 
 
 Oxygen used in 
 
 Alb. 
 ammonia 
 yielded. 
 
 5 minutes. 
 
 | hour. I 3 hours. 
 
 No. I. Fresh 
 
 9-49 
 8-22 
 
 13-91 
 11-80 
 10-30 
 
 26-08 
 20-74 
 18-99 
 2-02* 
 
 2-34* 
 
 13-52 
 8-00 
 5-42 
 
 o-io 
 
 . 
 0-21 
 
 2 days 
 
 No. II. 3 days 
 
 Edinburgh water ... 
 
 No. III. Ditto plus 20 c.c. 
 effluent II 
 
 
 * The oxygen stated is the quantity left after deducting the 
 quantity used up in five minutes from the total quantity used in 
 three hours. 
 
of Potable Waters 
 
 15 
 
 TABLE IY 
 Vegetable Solutions. 
 
 Strychnia I. A solution made by dissolving 100 
 milligrammes of pure crystals of strychnia in sul- 
 phuric acid, and diluting the solution up to 100 c.c. 
 with distilled water. 
 
 Strychnia II. Same strength as No. 1 solution. 
 
 Brucia. Solution made as above, and 1 c.c. con- 
 tained 1 milligramme of brucia. 
 
 Muriate of Morphia. Solution of same strength as 
 above. 
 
 Tea. Solution made by digesting 0*50 gramme of 
 tea in 100 c.c. of cold distilled water. At the end of 
 one hour the solution was filtered, and made up to 
 150 c.c. with distilled water. 
 
 Oxygen and ammonia are stated as parts per mil- 
 lion. 
 
 Substance. 
 
 Quantity 
 used. 
 
 Oxygeii used in 
 
 Ammonia 
 yielded. 
 
 \ hour. 
 
 3 hours. 
 
 Strychnia I 
 
 0-5 c.c. 
 
 ro 
 
 3-0 
 
 i-o 
 
 4-0 
 I'O 
 5-0 
 
 iio-o 
 
 101-0 
 265-0 
 377-0 
 341-0 
 67'5 
 
 54-0 
 
 150-8 
 119-3 
 489-0 
 432-0 
 559-0 
 142-5 
 
 86-6 | 
 
 72-0 
 70-0 
 58-0 
 35-0 
 33-3 
 36-0 
 Alb. am. 
 10-6 
 
 II .. 
 
 Brucia 
 Morph. mur 
 
 carefully dried... 
 Tea 
 
 12 days old 
 
 TTWTVEB.SITY 
 
16 
 
 The Organic Analysis 
 
 TABLE Y 
 Vegetable Solutions (continued). 
 
 Vegetable Albumen. Solution made by digesting 
 chips of raw potato in cold distilled water, and at the 
 end of one hour decanting off the water. 
 
 Wheaten Flour. Solution made by adding 100 
 milligrammes of the flour and 100 milligrammes of 
 caustic soda to distilled water, and making the solu- 
 tion up to 100 c.c. 
 
 Soluble Albumens of Wheaten Flour.4-Q-n.e g'ramme 
 of the same flour as above was digested in 100 c.c. of 
 distilled water, containing a few drops of liquor 
 potasses, for eighteen hours, and the solution was then 
 filtered. 
 
 Oxygen and ammonia are stated as parts per mil- 
 lion. 
 
 Substance. 
 
 Quantity 
 used. 
 
 Oxygen used in 
 
 Total 
 ammonia 
 yielded. 
 
 | hour. 
 
 3 hours. 
 
 Potato solution 
 
 10-0 c.c. 
 10-0 
 10-0 
 
 18-9 
 43-9 
 5-8 
 
 60-8 
 118-4 
 10-5 
 
 29-7 
 123-9 
 12-0 
 
 Flour albumens 
 
 Flour solution 
 
 
of Potable Waters 
 
 17 
 
 TABLE VI 
 Natural Peaty Waters. 
 
 Oxygen and albuminoid ammonia are stated as 
 parts per million, or milligrammes per litre. 
 
 Water. 
 
 Oxygen used in 
 
 Alb. 
 ammonia 
 yielded. 
 
 \ hour. 
 
 3 hours. 
 
 Edinburgh water from reservoir 
 on Alnwick Hill, not filtered 
 
 0-99 
 
 2-48 
 
 0-265 
 
 Loch Spalandar Old peat 
 
 5-56 
 
 10-69 
 
 0-215 
 
 2nd sample... 
 
 
 
 11-18 
 
 0-287 
 
 Patna Moor Surface stream ... 
 
 6-01 
 
 12-28 
 
 0-400 
 
18 
 
 The Organic Analysis 
 
 TABLE VII 
 Natural Waters, filtered previous to Use. 
 
 Oxygen and albuminoid ammonia are stated as parts 
 per million, or milligrammes per litre. 
 
 Date. 
 
 Water. 
 
 Oxygen used in 
 
 Alb. 
 ammonia 
 yielded. 
 
 \ hour. 
 
 3 hours. 
 
 Dec. 15, 1887 
 
 City supply, Edinburgh 
 
 1-02 
 
 2-24 0-187 
 
 16, 
 
 >j 
 
 0-88 
 
 2-20 
 
 0-183 
 
 Jan. 11, 1888 
 
 0-82 
 
 2-15 
 
 0-170 
 
 Feb. 8, 
 
 > 
 
 
 
 2-02 
 
 o-ioo 
 
 Mar. 12, 
 
 > 
 
 0-70 
 
 1-87 
 
 0-128 
 
 > 13, ,, 
 
 
 
 0-74 
 
 1-90 i 0-127 
 
 Apr. 17, 
 
 
 
 0-87 
 
 2-07 
 
 o-ioo 
 
 Nov. 22, 
 
 
 
 0-76 
 
 2-18 
 
 0-164 
 
 Dec. 7, 
 
 
 
 0-91 
 
 1-93 
 
 0-113 
 
 Jan. 14, 1889 
 
 M 
 
 0-86 
 
 1-66 
 
 0-098 
 
 Mar. 27, 
 
 t 
 
 0-64 
 
 1-62 
 
 0-126 
 
 Apr. 4, 
 
 " 
 
 0-64 
 
 1-46 
 
 0-132 
 
of Potable Waters 
 
 TABLE VIII 
 Well Waters. 
 
 Oxygen and albuminoid ammonia are stated as 
 parts per million, or milligrammes per litre. 
 
 Water. 
 
 Oxygen used in 
 
 Alh. 
 ammonia 
 yielded. 
 
 t> 
 ? 
 D 
 
 P 
 P 
 P 
 t> 
 D 
 S 
 
 i- 
 
 3 
 
 *r 
 ? 
 
 1 hour. 
 
 3 hours. 
 
 St Anthony's Well, Queen's Park . . . 
 
 0-14 
 000 
 0-12 
 0-04 
 009 
 0-34 
 1-42 
 0-26 
 0-46 
 0-21 
 0-03 
 0-28 
 0-05 
 0-06 
 0-03 
 
 o-oo 
 o-oo 
 o-oo 
 
 0-22 
 
 .0-00 
 
 . 0-29 
 , 0'13 
 ,0-26 
 0-89 
 2-72 
 0-42 
 1-01 
 0-45 
 0-13 
 0-51 
 0-17 
 
 0-11 
 
 0-25 
 009 
 0-12 
 000 
 
 0-115 
 0-014 
 0-127 
 0-052 
 0-077 
 0-315 
 0-355 
 0-055 
 0-116 
 0101 
 0-093 
 0-180 
 0053 
 0-025 
 0-066 
 0-063 
 0-024 
 0-017 
 
 
 St Margaret's Well, 
 
 
 t J> > 
 
 Manse Well, Straiten 
 
 
 
 " " 
 
 Surface Well, 
 
 TullWell, 
 
 
 " 
 
 
 Village Well, No. I, Straiten . . . 
 
 No. II, 
 
 Public School Water 
 
 Trough, Powburn 
 
 
20 
 
 The Organic Analysis 
 
 TABLE IX 
 Well Waters (continued). 
 
 Oxygen and albuminoid ammonia are stated as 
 parts per million, or milligrammes per litre. 
 
 Water. 
 
 Manse Well, Kirkmichael 0'18 
 
 Village 0-07 
 
 No. I 0-06 
 
 No. II 0-15 
 
 No. Ill 0-00 
 
 No. IV 0-00 
 
 No. I Patna O'OO 
 
 No. II 0-00 
 
 Peterhead Prison Supply 0'70 
 
 Sample II 0'92 
 
 Water No. I. Peterhead 0'56 
 
 II O'OO 
 
 III 0-97 
 
 IV 0-35 
 
 Spring. The Braids 0'02 
 
 Spout. Black ford Hill O06 
 
 Well. Northumberland Street, 0'18 
 
 Edinburgh 
 
 Oxygen used in 
 
 | hour. 
 
 3 hours. 
 
 0-2*4 
 0-20 
 0'12 
 0-30 
 0-25 
 0-10 
 0-12 
 
 o-oo 
 
 1-86 
 1-80 
 1-24 
 0-33 
 2-34 
 0-86 
 0-07 
 0-21 
 0-44 
 
 Alb. 
 
 ammonia 
 yielded. 
 
 0-037 
 
 0-090 
 
 0-037 
 
 0-110 
 
 0-060 
 
 0-040 
 
 0-035 
 
 0-025 
 
 0-170 
 
 0-152 
 
 0-100* 
 
 0-178* 
 
 0-194 
 
 0-120 
 
 0-035 
 
 0-042 
 
 0-128 
 
 * Much free ammonia, and in No. II a large quantity of ferrous 
 chloride. The oxygen used up by the ferrous salts is deducted in the 
 above table. 
 
of Potable Waters 
 
 21 
 
 TABLE X 
 Polluted Waters. 
 
 Containing much organic matter of animal and 
 vegetable origin. 
 
 Oxygen and albuminoid ammonia are stated as 
 parts per millioD, or milligrammes per litre. 
 
 Water. 
 
 Oxygen 
 i Lour. 
 
 used in 
 3 hours. 
 
 Alb. 
 ammonia 
 yielded. 
 
 St Margaret's Loch, Queen's Pk. 
 
 0-66 
 
 1-39 
 
 0-700 
 
 5> 
 
 0-69 
 
 1-56 
 
 0-485 , 
 
 J> 
 
 094 
 
 1-96 
 
 1-247 
 
 5> J 
 
 0-88 
 
 1-43 
 
 0-380 
 
 Dunsappy 
 
 0-58 
 
 1-44 
 
 0-550 
 
 )t 
 
 0-80 
 
 1-59 
 
 0-245 
 
 Duddingston 
 
 1-21 
 
 2-42 
 
 0-367 
 
 }J 
 
 1-23 
 
 2-70 
 
 0-350 
 
 j> 
 
 1-26 
 
 2-98 
 
 0-277 
 
 River Esk, Hawthornden 
 
 2'88 
 
 5-92 
 
 0-730 
 
 same sample 
 
 
 5-55 
 
 0-730 
 
 Water of Leitb, Colinton Bridge 
 
 3-60 
 
 6-97 
 
 1-050 
 
 same sample ... 
 
 3-75 
 
 6-90 
 
 0-765 
 
 Dean Bridge ... 
 
 243 
 
 7-16 
 
 0-505 
 
 j> 
 
 2-29 
 
 6-92 
 
 0-466 
 
 Bonnington ... 
 
 5-44 
 
 18-90 
 
 2-720 
 
 )> 
 
 3-47 
 
 13-59 
 
 2-060 
 
 
 
 3-89 
 
 15-13 
 
 2-280 
 
22 The Organic Analysis 
 
 The organic impurities of potable waters are divided, 
 for practical purposes, into two divisions : 
 
 1. Organic substances of animal origin. 
 
 2. Organic substances of vegetable origin. 
 
 The first group of organic bodies is regarded as 
 much more dangerous to health than the latter. Fluids 
 containing organic bodies of animal origin often carry 
 the germs of diseases associated with the presence of 
 animals, and for this reason it is of great importance 
 that, on analysis, the organic matters in a water can 
 be assigned to their proper class. Unfortunately, 
 when the quantity of organic matter in* a sample of 
 water is very small, it is not always possible to do 
 this, and yet it is just in these cases that the question 
 is so important. When the quantity of organic 
 matter is large, the water is condemned irrespective 
 of the origin of the matter. 
 
 The series of analyses, contained in the tables 
 alluded to, was made with the object of showing the 
 results got when examining the same organic solution 
 by the albuminoid ammonia and oxygen processes. 
 In order to facilitate the comparison between the two 
 methods, the results in both cases are stated as parts 
 per million. When very small quantities of the 
 organic solutions were used, the alkaline perman- 
 ganate was boiled in a distillation flask with distilled 
 water till the distillate gave no colour reaction with 
 
of Potable Waters 23 
 
 Nessler's reagent, and then the organic fluid was 
 dropped into the permanganate solution. 
 
 Solutions of Organic Matter of Animal Origin. 
 
 Table I shows that fresh white of egg yields up as 
 much ammonia as it uses up of oxygen in three hours, 
 and that, in all cases, the oxygen used up in a quarter- 
 hour is considerably less than the quantity of am- 
 monia yielded. 
 
 Table II shows that fresh urine yields up consider- 
 ably more ammonia than it uses up of oxygen in 
 three hours, and the difference is all the more marked 
 if compared with the oxygen used in a quarter-hour. 
 The table also shows that these results hold good 
 when urine is mixed with a natural water, and let 
 stand for twelve days before analysing. 
 
 Table III shows that effluent waters from sewage 
 farms, when not thoroughly purified, contain organic 
 substances in a very readily oxidisable condition, and 
 consequently the oxygen used up in three hours is, at 
 least, twice as much as the albuminoid ammonia 
 yielded in some cases the ratio is even greater. The 
 effluent waters examined doubtless contained organic 
 substances of vegetable as well as of animal origin. 
 The oxygen used in a quarter-hour is about equal to 
 the albuminoid ammonia yielded. The ratio of oxygen 
 
24 The Organic Analysis 
 
 to ammonia remained unchanged when the effluent 
 water was added to a sample of the water sup plied, to 
 Edinburgh city. 
 
 Solutions of Organic Matter of Vegetable Origin. 
 
 Table IV shows that the vegetable alkaloids expe- 
 rimented with use up far more oxygen in three hours 
 than ammonia yielded, the ratio of oxygen to am- 
 monia being roughly as 2 to 1, and in some cases 
 much greater. The oxygen used up in a quarter-hour, 
 in all cases, considerably exceeded the ammonia 
 yielded. The same remarks hold good in the case 
 of tea. 
 
 Table V shows the experimental results obtained 
 from vegetable albumens. The solution of albumen 
 from potato resembles the effluent waters examined, 
 in that the oxygen used in three hours has a ratio of 
 2 to 1 to the ammonia yielded. The albuminoid solu- 
 tions from wheaten flour gave the same ratio of oxy- 
 gen used to ammonia yielded as in the case of egg 
 albumen. 
 
 Table VI shows the oxygen used in three hours by 
 peaty waters is greatly in excess of the albuminoid 
 ammonia yielded, the ratio ranging in the samples 
 examined from 10 to 1 up to 50 to 1. The oxygen 
 absorbed in a quarter-hour exceeds the ammonia 
 
of Potable Waters 25 
 
 yielded in a ratio ranging from 4 to 1 up to 25 to 1. 
 The ratio seems to be highest in waters from old 
 peat. 
 
 Table VII shows that when peaty waters are filtered 
 -the oxygen absorbed in three hours still exceeds the 
 .ammonia yielded in a ratio of 25 to 1. 
 
 Tables VIII and IX contain waters which are nearly 
 all of great purity in some five or six cases the waters 
 are suspicious. In all cases, the oxygen absorbed in 
 three hours exceeds the albuminoid ammonia yielded, 
 except in cases where the water is of extraordinary 
 organic purity. In most cases, the oxygen is in con- 
 .siderable excess over the ammonia, and in a few cases 
 the ratio is as 9 to 1. The oxygen absorbed in a 
 quarter-hour in some cases exceeds the albuminoid 
 ammonia, and in other cases it is less. 
 
 Table X. In this series of polluted waters the oxy- 
 gen absorbed in three hours exceeds the albuminoid 
 :ainmonia in a ratio ranging from 1J to 1 up to about 
 10 to 1. The oxygen absorbed in a quarter-hour in 
 all cases, at least, equals the albuminoid ammonia, and 
 generally it is greatly in excess of the ammonia. 
 
 Summary of Results obtained from the Tables. 
 
 I. Solutions of egg albumen and of urine yield as 
 much as, or more albuminoid ammonia than they ab- 
 
26 The Organic Analysis 
 
 sorb of oxygen in three hoars. This is also true of 
 the vegetable albumen of wheat en flour, but in the 
 case of the albumen from potato, the oxygen absorbed 
 in three hours is about twice as much as the albumin- 
 oid ammonia yielded. 
 
 II. Effluent waters contain the organic matter in a 
 highly oxidisable condition, and much of the albu- 
 minoid ammonia has been converted into free ammonia, 
 consequently the oxygen used up in three hours was 
 found to be one and a half times to twice as much as 
 the albuminoid ammonia, even whe^i the oxygen used 
 by the nitrites is deducted from the total quantity used. 
 If the oxygen used in five minutes (due to nitrites) is 
 deducted from the oxygen absorbed in a quarter- 
 hour, it is found that the albuminoid ammonia exceeds 
 the oxygen used in a quarter-hour in a ratio of 2 
 to 1 up to 3-^ to 1. 
 
 III. Well waters containing only vegetable organic 
 matter, as a rule, absorb much more oxygen in three 
 hours than they yield of albuminoid ammonia. This 
 may be due to the waters containing oxidisable sub- 
 stances, which do not yield ammonia, such as starch, 
 gum, sugar, and vegetable alkaloids containing little 
 nitrogen. Complete oxidation of albumen requires 
 nearly ten times as much oxygen as it is capable of 
 yielding of ammonia. 
 
 IV. Vegetable alkaloids usually contain little nitro- 
 gen compared with the large quantity of carbon, con- 
 
of Potable Waters 27 
 
 sequently the oxygen absorbed in three hours exceeds 
 the ammonia yielded. 
 
 V. Polluted waters, containing both vegetable and 
 animal organic matters. If Edinburgh sewage, as it 
 enters Craigentinny Meadows, be taken as a type of 
 sewage waters in general, it is evident that a natural 
 water when polluted with sewage water will still ab- 
 sorb much more oxygen in three hours than it will 
 yield up albuminoid ammonia, because sewage water 
 itself has its oxygen absorbed in three hours in excess 
 of the albuminoid ammonia at least in a ratio of 1-j- to 1. 
 Of course such a polluted water will contain much 
 " free ammonia/' Edinburgh water, in which the 
 oxygen absorbed in three hours exceeded the albu- 
 minoid ammonia yielded in the ratio of 20 to 1, after 
 the addition of 20 c.c. of the Craigentinny effluent had 
 this ratio reduced to about 11 to 1 . 
 
 VI. Peaty waters use very large quantities of oxy- 
 gen in three hours compared with the albuminoid am- 
 monia yielded, the ratio of oxygen to albuminoid 
 ammonia ranging from 10 to 1 up to 50 to 1, or even 
 greater. 
 
 The following is Mr Wanklyn's classification of 
 waters, based upon his albuminoid ammonia pro- 
 cess. 
 
The Organic Analysis 
 
 pof great purity if the alb. ammonia is less 
 than 0*05 part per million. 
 
 organically safe if the alb. ammonia is less 
 A water is < , 1 .... 
 
 than O'lO part per million. 
 
 a dirty water if the alb. ammonia is more 
 _ than 0*10 part per million. 
 
 Professor Tidy's classification of waters, based upon 
 the oxygen used, is 
 
 f of great organic purity if the oxygen used 
 in three hours is less than 0'05 part 
 per 100,000. 
 
 of medium organic purity if the oxygen 
 A water is <( use a j n three hours is from 0'05 to 0'15 
 
 part per 100,000. 
 
 of suspicious organic purity if the oxygen 
 used in three hours is from 0'15 to 0'21 
 part per 100,000. 
 
 I 
 
 I believe that the range of suspicious waters is ex- 
 tended up to 0*30 part of oxygen per 100,000, and 
 above this point waters are considered as unfit for 
 
 domestic use. 
 
 
of Potable Waters 
 
 The Advantages of examining Samples of Waters by both 
 the Albuminoid Ammonia and Oxygen Processes. 
 
 If one process be trusted to only, errors of judgment 
 may be made, and a water may be condemned as 
 unfit for use when it is really a drinkable water. Of 
 the well waters examined (see Tables VIII and IX) 
 eight of them would be pronounced suspicious, or 
 even dirty waters in some cases, if the albuminoid 
 ammonia process had been alone trusted to ; if these 
 eight waters had been analysed only according to the 
 oxygen process most of them would have been pro- 
 nounced as waters of great purity, and all of them 
 stated to be potable waters. 
 
 On the other hand, peaty waters are liable to be 
 condemned as suspicious or dirty waters by the oxygen 
 process, when pronounced potable according to Mr 
 Wanklyn's classification. 
 
 It is evident that these two classifications of waters 
 are not in harmony with one another. This state of 
 affairs has led to contradictory reports being made 
 upon the same water, where the analysts had relied 
 on one process only for estimating the organic matter. 
 
 If a water is not condemned by much "free am- 
 monia " or nitrites, and if the chlorine is moderate in 
 quantity, the above experiments seem to indicate the 
 following conclusions : 
 
30 The Organic Analysis 
 
 (1) If the oxygen absorbed in three hours does not 
 exceed O05 part per 100,000 the quantity of albumin- 
 oid ammonia yielded is of little consequence as long 
 as it does not exceed 0*20 part per million. 
 
 (2) When the oxygen absorbed in three hours is as 
 high as O'lO part per 100,000 the albuminoid ammonia 
 should be at least ten times less, or not more than 0*12 
 part per million. 
 
 (3) If the oxygen absorbed in three hours rises as 
 high as 0*15 part per 100,000 the albuminoid ammonia 
 should be at least twelve times less, or not more than 
 0*12 part per million. 
 
 (4) When the oxygen absorbed in three hours is 
 more than 0'15 part per 100,000 the albuminoid 
 ammonia should be at least fifteen times less, or not 
 more than O'lO part per million, and running up to 
 0*20 part per million in proportion as the quantity 
 of oxygen absorbed rises above 0'15 part per 100,000. 
 Waters absorbing so much oxygen in three hours can 
 only be safe if the organic matter is peaty, and in 
 these cases the ratio of oxygen to ammonia is natur- 
 ally high. 
 
 If a water fulfils these conditions it may be regarded 
 as a potable water, as far as the organic matter is 
 concerned. 
 
of Potable Waters 31 
 
 PAET II 
 
 THE OXYGEN PROCESS AT 100 C. 
 
 THE albuminoid ammonia and oxygen processes are 
 faulty in several ways. 
 
 The chief objection to the oxygen process at ordin- 
 ary temperatures is 
 
 By the oxygen process, as usually carried out, the 
 organic matter is only partially oxidised ; at best the 
 oxygen absorbed in three hours is only about a quarter 
 of the oxygen required for complete oxidation of the 
 organic matter, and in many cases it is a much smaller 
 fraction of the total quantity required. The albumin- 
 oid solution from wheaten flour, and the solution of 
 moist white of egg only absorbed in three hours about 
 a twenty -fifth of the oxygen required for the complete 
 oxidation of the organic matter in the solutions, and, 
 even under the most favorable conditions, the albumin- 
 oid solution from potato only absorbed in three hours 
 a little more than one-fifth of the oxygen required for 
 
32 The Organic Analysis 
 
 complete oxidation. During three hours, at ordinary 
 temperatures, starch and cane sugar absorb almost no 
 oxygen compared with the quantity required for the 
 complete oxidation of these substances. The proof of 
 these statements will be shown later on, in a series of 
 tables. 
 
 After the consideration of this objection to the 
 oxygen process as usually carried out, the question 
 naturally arises Can this objection to the oxygen 
 process be overcome or minimised ? 
 
 Hitherto, the best method of carrying out the oxygen 
 process is the one recommended by Professor Tidy, 
 the working detaijs of which were published in the 
 ' Chemical Society's Journal ' for 1879. Since then his 
 plan has usually been followed, some analysts modi- 
 fying the time of exposure of the water to the action 
 of the oxidising agents, and others modifying the 
 temperature. It has been suggested to keep the 
 water in stoppered bottles at a temperature of 80 F. 
 for four hours, whilst it is exposed to the action of the 
 permanganate. If the quantity of oxygen absorbed 
 by a water in a given time can be used as a reliable 
 basis for classifying the organic purity of waters, even 
 when this quantity of oxygen is only a small fraction 
 of the total quantity the organic matter is capable of 
 absorbing, it is reasonable to suppose that the nearer 
 the quantity absorbed can be made to approach the 
 quantity required the more satisfactory will be the 
 
of Potable Waters 33 
 
 classification. The oxygen process is simply used as 
 a means of roughly determining, in the absence of 
 nitrites, ferrous salts, and sulphuretted hydrogen, the 
 relative quantity of organic matter present in a potable 
 water; but were it proved that the deoxidising property 
 of organic matter is per se hurtful to the human body, 
 then it is evident that the temperature of the human 
 body would be the proper one at which to carry out 
 the oxygen process. In the absence of this proof, the 
 temperature of the water during the process of analysis 
 is not of vital importance. Heat is a great promoter 
 of chemical action, and the oxidation of the organic 
 matter is much more complete if the mixture of 
 water, acid, and permanganate is boiled for some 
 time. 
 
 It is necessary to prove that the results obtained by 
 boiling in this way can be relied upon. 
 
 In order to furnish this proof, the following points 
 must be considered. 
 
 (1) Is a solution of permanganate of potash at 100 
 C. stable in the absence of reducing agents ? 
 
 (2) If stable at this temperature, is this stability 
 affected by the addition of sulphuric acid or alkali to 
 the solution ? 
 
 (3) Does the reducing power of organic solutions 
 vary with the quantity of acid or alkali present in the 
 solution, when the quantity of acid or alkali varies 
 within moderate limits ? 
 
 3 
 
34 The Organic Analysis 
 
 (4) Does concentration of the solution affect the 
 stability of a permanganate solution ? 
 
 (5) Does concentration of the permanganate solution 
 during the boiling process promote the oxidation of 
 organic substances ? 
 
 (6) If the answers to these questions are favorable 
 to the use of a temperature of 100 C. for the oxygen 
 process, what is the best way* of carrying out the 
 operation ? 
 
 (7) How do the results compare with those obtained 
 when the ordinary temperature is employed, and what 
 conclusions may be drawn from the working of the 
 process ? 
 
 The points will be considered in the above order, 
 and tables have been drawn up containing the experi- 
 ments made to clear up these various points. 
 
 Table XI furnishes the answer to the first question 
 by showing that a solution of permanganate of potash 
 can be boiled in pure distilled water for one hour 
 without suffering any reduction. 
 
 In all the experiments contained in the tables 
 dealing with the various questions under consideration, 
 a 24-oz. flat-bottomed glass flask was used, and the 
 mouth of the flask simply covered by a wide glass 
 funnel. 
 
 Table XII supplies the answer to the second ques- 
 tion by showing that a permanganate solution con- 
 taining 2*5 c.c. pure sulphuric acid can be boiled for 
 
of Potable Waters 35 
 
 two hours without suffering any reduction. After 
 this length of time there is a slight reduction of per- 
 manganate, varying with the duration of boiling. 
 This reduction seems to depend upon the concentra- 
 tion of the solution by loss from evaporation, as half 
 a litre of water will stand boiling for two hours with- 
 out any reduction of permanganate, whereas a quarter 
 of a litre of water at the end of two hours shows a 
 slight reduction of permanganate. Possibly the ex- 
 planation of this is that, as the volume of water dimin- 
 ishes by evaporation, some of the permanganate 
 adheres to the flask above the water-line, and is then 
 reduced to a lower oxide of manganese. 
 
 Table XIII is not favorable to the use of alkaline 
 solutions of permanganate for estimating the quantity 
 of oxygen absorbed. The addition of caustic potash 
 or soda caused a reduction of permanganate, even when 
 the solution was only boiled for one hour. The quan- 
 tity of oxygen used up in these blank experiments 
 roughly varied with the quantity of caustic alkali 
 used. This slight reduction of permanganate may be 
 due to the caustic alkali in stick containing sulphur 
 compounds, which act as reducing agents. 
 
 Table XIV furnishes the answer to the third ques- 
 tion by showing that Edinburgh water practically ab- 
 sorbed the same quantity of oxygen, when the quanti- 
 ties of dilute sulphuric acid varied from 1 c.c. up to 
 20 c.c. It is evident that, when half a litre of water 
 
36 The Organic Analysis 
 
 is used, there is no advantage in using large quantities 
 of sulphuric acid. 
 
 The fourth question is answered by Table XII and 
 the remarks made upon it. There is no reduction of 
 permanganate during evaporation, if the concentra- 
 tion is kept within certain limits. A quarter of a litre 
 of permanganate solution may be boiled for one hour, 
 and half a litre for two hours, without suffering any 
 reduction of permanganate, if certain precautions are 
 observed. (See working details of the oxygen process 
 at 100 C.) Eoughly speaking, an acid solution of 
 permanganate can be reduced to half its original 
 volume by evaporation, whilst boiling, without suffer- 
 ing any reduction of permanganate. 
 
 Table XV answers the fifth question by showing 
 that oxidation of the organic matter in Edinburgh 
 water is not hastened by allowing a J litre of the 
 water to be boiled down to 60 c.e., if the loss of per- 
 manganate due to concentration the quantity due to 
 this cause being experimentally found be deducted 
 from the total quantity reduced. 
 
 The answers to these five questions are favorable to 
 the use of a temperature of 100 C. for the oxygen 
 process, if certain precautions are observed. 
 
of Potable Waters 37 
 
 Working Details for the Oxygen Process at 100 C. 
 
 Experiments prove that a solution of permanganate 
 and sulphuric acid can be boiled for two hours with- 
 out any reduction of permanganate, if reducing agents 
 are absent, and I shall now describe what I have found 
 to be the best way of carrying out the process. 
 
 It is desirable to use, at least, half a litre of the 
 sample of water, or if a smaller quantity must be used, 
 it ought to be made up to half a litre with pure dis- 
 tilled water. For this purpose, organically pure dis- 
 tilled water is prepared from a good potable water 
 containing excess of permanganate in solution. It is 
 not necessary to use a litre of the sample of water, as 
 the quantity is too bulky to manage readily, and the 
 experimental error with half a litre of water should be 
 so small as not to affect general results. 
 
 Solutions required : 
 
 Solution of permanganate of potash. 
 
 Dilute solution of pure sulphuric acid. 
 
 Solution of potassium iodide. 
 
 Solution of sodium hyposulphite. 
 
 Solution of starch. 
 
 The preparation of these solutions is given in 
 Appendix C. Thoroughly cleanse a 24-oz., flat- 
 bottomed flask with strong sulphuric acid, then rinse 
 out well with tap water, and lastly with a small quan- 
 
38 The Organic Analysis 
 
 tity of distilled water. Into this flask pour half a 
 litre of the sample of water, add 10 c.c. of the per- 
 manganate solution and 10 c.c. of the dilute sulphuric 
 acid j place a clean, moderately sized glass funnel into 
 the neck of the flask, and stand the flask on a sheet of 
 wire-gauze, covering the ijing of an iron stand. It is 
 advisable to put several small pieces of platinum-foil 
 into the flask, so as to avoid bumping of the acid 
 liquid. Bring the water to the boil, and then note the 
 time ; continue to boil gently for two hours over the 
 wire-gauze, or transfer the flask to a sand-bath, where 
 the water will boil gently without bumping, as the 
 heat is thus more evenly applied to the flask. The 
 more gently the liquid is boiled, the less will be the 
 loss from evaporation a point to be kept in view. 
 It is best to use a glass funnel from which the stem 
 has been filed off, so as to allow the watery vapour to 
 escape easily ; otherwise, the vapour will pass off at in- 
 tervals with explosive violence, and thereby may cause 
 partial loss of the liquid contents of the flask. 
 
 The water should be boiled for two hours, and, dur- 
 ing this time, the loss from evaporation should not be 
 more than a quarter, or a third, of the original volume ; 
 this will ensure that there is no reduction of per- 
 manganate due to a too great concentration of the 
 liquid. 
 
 During the whole time the water is boiling, it should 
 remain decidedly pink when the flask is held against 
 
of Potable Waters 39 
 
 the light ; if the pink colour becomes very faint, 
 another 10 c.c. of the permanganate solution must be 
 added. 
 
 At the end of two hours, remove the flask from the 
 sand-bath and allow the water to cool for ten minutes ; 
 then hasten the cooling by placing the flask in a basin 
 of cold water, or by holding it under a stream of cold 
 water running from the tap. When the water is 
 reduced to the ordinary temperature, add 20 drops of 
 the solution of potassium iodide for every 10 c.c. of 
 the permanganate solution used. Agitate gently the 
 contents of the flask, and the precipitate of manganic 
 and manganese oxides will soon disappear, and give 
 place to a clear, brilliant, yellow-brown colour. Run 
 in the hyposulphite solution from a burette till the 
 water is nearly colourless, then add enough of the 
 solution of starch to render the solution a deep blue 
 in colour, and continue to add the hyposulphite solu- 
 tion till the blue colour just disappears. The hypo- 
 sulphite solution is standardised by adding 10 c.c. of 
 the above potassium permanganate solution to J litre 
 of distilled water, also 10 c.c. of dilute sulphuric acid, 
 and 20 drops of the solution of potassium iodide, and 
 proceeding exactly in the way already described. 
 
 The quantity of oxygen absorbed by the organic 
 matter is calculated in the same way as for the 
 ordinary oxygen process, keeping in mind that 10 
 c.c. of the solution of permanganate are capable of 
 
40 The Organic Analysis 
 
 yielding up 5 milligrms. of oxygen for oxidation 
 purposes. 
 
 It is "better to add sulphuric acid to the sample of 
 water than to use permanganate alone, because 
 sulphuric acid hastens the oxidation of organic matter. 
 
 For example, 20 milligrms. o,f wheaten starch, when 
 boiled for two hours with sulphuric acid and perman- 
 ganate solution, absorbed 17*28 milligrms. of oxygen; 
 whereas, the same quantity, when boiled for the same 
 time with permanganate solution without the addi- 
 tion of the acid, only absorbed 5*58 milligrms. of 
 oxygen. 
 
 Irrespective of organic matter, the reducing agents 
 likely to be present in a water and to reduce perman- 
 ganate are (1) chlorides, (2) nitrites, (3) ferrous 
 salts, and (4) sulphuretted hydrogen. 
 
 The sulphuric acid will partially decompose the 
 chlorides, forming hydrochloric acid. The quantity 
 of hydrochloric acid formed in this way is very small, 
 and becomes highly diluted in the J litre of water. 
 Experimentally, it was found that a solution contain- 
 ing eight grains of chlorine per gallon practically 
 reduced no permanganate through the formation of 
 hydrochloric acid. Comparatively few waters con- 
 tain so much chlorine, so the reduction of perman- 
 ganate due to chlorides may be disregarded. 
 
 Nitrites and Ferrous Salts. The oxygen absorbed, 
 due to the presence of either of these compounds, may 
 
of Potable Waters 41 
 
 be estimated at the end of five minutes (see Appen- 
 dix C). 
 
 Sulphuretted hydrogen, or nitrous acid, if present, 
 can be boiled off after the addition of the sulphuric 
 acid, before the permanganate is added. 
 
 The quantity of oxygen absorbed by the nitrites 
 and ferrous salts must be deducted from the total 
 quantity of oxygen absorbed, in order to find the 
 oxygen absorbed by the organic matter. 
 
42 
 
 The Organic Analysis 
 
 TABLE XI 
 
 Experiments with a Solution of Permanganate of 
 Potash in organically Pure Distilled Water. 
 
 Distilled water was the distillate from Edinburgh 
 City water, containing excess of permanganate in 
 solution. 
 
 Solution of permanganate 10 c.c., capable of 
 yielding up 5 milligrms. of oxygen. 
 
 Solution was kept at the boiling-point. 
 
 Capacity of flask was 24 oz. Glass funnel used. 
 
 Quantity of aqua 
 
 Quantity of 
 permang. solution. 
 
 Duration of 
 boiling. 
 
 Oxygen used. 
 
 I. i litre 
 II. * 
 
 10 c.c. 
 10 c.c. 
 
 1 hour 
 
 O'OO milligrm. 
 
 o-oo 
 
of Potable Waters 
 
 43 
 
 TABLE XII 
 
 Experiments with Sulphuric Acid and a Solution of 
 Permanganate of Potash in organically Pure Dis- 
 tilled Water. 
 
 In all cases 10 c.c. of a 1 in 4 solution of pure sul- 
 phuric acid were used. 
 
 The experiments were conducted under the same 
 conditions as in Table XI. (See Strength of Solu- 
 tions, &c.) 
 
 Quantity of aqua 
 destillata. 
 
 Quantity of 
 permnng. solution. 
 
 Duration of 
 boiling. 
 
 Oxygen absorbed. 
 
 I. * litre 
 
 10 c.c. 
 
 1 hour 
 
 O'OO milligrm. 
 
 II- * n 
 
 20 c.c. 
 
 1 
 
 o-oo 
 
 HI. i 
 
 10 c.c. 
 
 1 
 
 o-oo 
 
 IV. * 
 
 20 c.c. 
 
 2 hours 
 
 o-oo 
 
 v. * 
 
 10 c.c. 
 
 2 
 
 0-02 
 
 VI. i 
 
 10 c.c. 
 
 2 
 
 0-22 
 
 VII. t 
 
 10 c.c. 
 
 2* 
 
 019 
 
 VIII. 1 
 
 10 c.c. 
 
 3 
 
 0-22 
 
 IX. i 
 
 20 c.c. 
 
 3 
 
 0-12 
 
 X. i 
 
 10 c.c. 
 
 3 
 
 0-40 
 
 XL i 
 
 20 c.c. 
 
 5 
 
 0-31 
 
44 
 
 The Organic Analysis 
 
 TABLE XIII 
 
 Experiments with an Alkaline Solution of Permanganate 
 of Potash in organically Pure Distilled Water. 
 
 Strength of permanganate solution, &c., the same 
 as in Table XI. 
 
 In each experiment 10 c.c. of permanganate solution 
 were used. 
 
 The caustic alkali used was in stick. 
 
 Quantity of aqua 
 destillata. 
 
 Quantity of alkali. 
 
 Duration of 
 boiling. 
 
 Oxygen absorbed. 
 
 I. 400 c.c. 
 
 1 grm. caustic potash 
 
 1 hour 
 
 0-12 milligrm. 
 
 II. 500 c.c. 
 
 2 grms. 
 
 1 
 
 0-28 
 
 HI. 
 
 5 
 
 1 
 
 0-45 
 
 IV. 
 
 4 caustic soda 
 
 1 
 
 0-36 
 
 Note. In Experiments II and III ordinary distilled 
 water was used, and 0*228 milligrm. of oxygen 
 was deducted from the quantity of oxygen absorbed, 
 as this was the quantity of oxygen absorbed by the 
 distilled water containing sulphuric acid and perman- 
 ganate. 
 
of Potable Waters 
 
 45 
 
 TABLE XIY 
 
 Experiments showing that the Oxidising Power of a 
 Solution of Permanganate is not affected by Varia- 
 tions of the Acidity of the Solution within Moderate 
 Limits. 
 
 Conditions of experiments the same as in Table XI. 
 In each experiment the duration of boiling was one 
 hour. 
 
 Date. 
 
 Quantity of 
 water. 
 
 Quantity of 
 acid. 
 
 Quantity of 
 perniang. solution. 
 
 Oxygen absorbed. 
 
 1888 
 
 
 
 
 
 Nov. 20 
 
 litre Edinb. 
 
 1 C.C. 
 
 10 c.c. 
 
 2-91 milligrms. 
 
 
 water 
 
 
 
 
 20 
 
 > 
 
 2 c.c. 
 
 
 2-88 
 
 21 
 
 
 2 c.c. 
 
 
 2-91 
 
 21 
 
 > j 
 
 5 c.c. 
 
 
 2-99 
 
 22 
 
 
 5 c.c. 
 
 
 2-87 
 
 Dec. 15 
 
 4 > 
 
 5 c.c. 
 
 
 1-60 
 
 ,, 15 
 
 J> JJ 
 
 20 c.c. 
 
 
 1-63 
 
46 
 
 The Organic Analysis 
 
 TABLE XV 
 
 Experiments showing that Concentration of an Organic 
 Solution does not hasten Oxidation in the Presence 
 
 of Sulphuric Acid and Permanganate. 
 
 * 
 
 For these experiments it was convenient to use 
 Edinburgh City water, because it is a peaty water, 
 and, beyond a certain point, its organic matter is only 
 slowly oxidised in the heat by acid and permanganate. 
 Possibly the cellulose in peaty waters is only slightly 
 oxidised by permanganate. 
 
 Strength of solutions the same as in Table XI. 
 For every quarter-litre of water used 10 c.c. of per- 
 manganate solution were added, therefore when half a 
 litre was used 20 c.c. of permanganate solution were 
 added. 
 
 Date. 
 
 Quantity of 
 water. 
 
 Volume 
 reduced to 
 
 Duration 
 of boiling. 
 
 Oxygen 
 absorbed. 
 
 Oxygen used 
 per I litre. 
 
 1888. 
 
 
 
 
 Millig. 
 
 
 Dec. 20 
 
 i litre aq. dest. 
 
 60 c.c. 
 
 1 hour 
 
 0-28 
 
 
 
 20 
 
 Edin.water 
 
 
 li hours 
 
 1-73 
 
 (1-730-28) 
 
 
 
 
 
 
 = 1-45 millig. 
 
 18 
 
 i litre 
 
 Scarcely 
 
 1 hour 
 
 2-85 
 
 1-42 
 
 
 
 reduced 
 
 
 
 
 19 
 
 " 
 
 
 
 1 
 
 3-02 
 
 1-51 
 
 Note. Distilled water prepared as in Table XI 
 
of Potable Waters 
 
 47 
 
 does not reduce permanganate, therefore 0*28 milli- 
 grm. oxygen absorbed in the above experiment is 
 due to concentration, and must be deducted from the 
 oxygen absorbed by Edinburgh water in the above 
 experiment. 
 
 UNIVERSITY 
 
48 
 
 The Organic Analysis 
 
 TABLE XVI 
 
 Table showing the Quantities of Oxygen absorbed by the 
 same Organic Solutions at Ordinary Temperatures, 
 and at 100 C. 
 
 The quantity of the solution was made up to half 
 a litre with organically pure distilled water. 
 
 In all experiments 10 c.c. of a 1 in 4 solution of pure 
 sulphuric acid were used. 
 
 Excess of permanganate solution was added in all 
 cases. 
 
 The oxygen is stated as so much per cent. 
 
 Substance, 100 milligrms. 
 
 Ordinary temp. 
 
 Temp. = 100 C. 
 
 Oxygen 
 required 
 (calcu- 
 lated). 
 
 Oxygen 
 i 
 
 ( 
 
 \ hour. 
 
 absorbed 
 
 a 
 
 A ^ 
 
 3 hours. 
 
 Oxygen 
 i 
 
 fhour. 
 
 absorbed 
 
 
 
 2 hours. 
 
 Cane su^nr ... 
 
 o-o 
 o-o 
 
 1-9 
 
 4-4 
 0-5 
 0-5 
 9-2 
 26-5 
 37-7 
 0-1 
 
 o-o 
 
 0-2 
 
 6-1 
 6-2 
 
 5-9 
 1-0 
 0-7 
 14-1 
 48-9 
 43-2 
 0-2 
 
 97-4 
 31-6 
 
 63-2 
 70-3 
 6-7 
 214-0 
 184-9 
 173-6 
 
 ri 
 
 0-9 
 
 111-2 
 
 86-4 
 
 67-8 
 
 88-8 
 84-2 
 8-2 
 
 1-1 
 
 111-9 
 118-5 
 
 32-1* 
 67-3* 
 
 143-2* 
 17-9* 
 
 Starch 
 
 I. Albuminoid solution from 
 potato 
 
 II. Ditto 
 
 Albuminoid solution from 
 wheaten flour ...... 
 
 Wheaten flour 
 
 
 
 Brucia 
 
 Muriate of morphia 
 
 Ditto 
 
 
 * Total solids are reckoned as albumen only approximately cor- 
 rect and the nitrogen was found by the sulphuric acid process. 
 
of Potable Waters 
 
 49 
 
 TABLE XYII 
 
 Table showing the Quantities of Oxygen absorbed by 
 Natural Waters at Ordinary Temperatures, and at 
 100 C. 
 
 The two oxygen processes were carried out in the 
 way already described. 
 
 The oxygen is stated as parts per million. 
 
 Water. 
 
 Ordinary temp. 
 
 Temp. 
 =100 C. 
 
 Ratio of 
 oxygen in 
 3 hours to 
 
 ThSs" 
 
 Oxygen a 
 
 I ' 
 | hour. 
 
 >sorbed in 
 3 hoars. 
 
 Oxygen 
 absorbed in 
 2 hours. 
 
 Edinburgh City supply 
 
 0-64 
 0-64 
 0-97 
 0-35 
 0-06 
 0-18 
 0-07 
 0-15 
 
 o-oo 
 o-oo 
 o-oo 
 o-oo 
 
 5-44 
 3-47 
 0-61 
 3-89 
 
 1-62 
 1-46 
 2-24 
 0-86 
 0-12 
 0-24 
 0-20 
 0-30 
 0-25 
 0-10 
 
 o-oo 
 
 0-12 
 18-90 
 13-59 
 3-37 
 15-13 
 
 6-13 
 5-72 
 9-12 
 4-28 
 0-90 
 0-61 
 1-37 
 2-31 
 1-54 
 0-88 
 0-50 
 0-73 
 101-25 
 107-45 
 26-67 
 108-05 
 
 3-8 
 3-9 
 4-0 
 4-9 
 7-5 
 2-5 
 
 6-8 
 
 7.7 
 
 6-1 
 
 8-1 
 
 6-0 
 5-3 
 
 7-8 
 7-9 
 
 7-2 
 
 I. From Pcterhead 
 
 II. 
 
 I. Well, Kirkmichael 
 II. ,, 
 HI. 
 IV. ,, 
 V. 
 VI. 
 I. Spring, Patna 
 
 II. Well 
 
 Sewage water 
 
 
 purified by alum 
 before filtration 
 
 4 
 
50 
 
 The Organic Analysis 
 
 TABLE XVIII 
 
 Table showing the Quantities of Oxygen absorbed by 
 Waters at the Ordinary Temperature for Three 
 Hours, and at 100 C. 
 
 Oxygen is stated as parts per million. 
 
 Water. 
 
 Oxygen absorbed in 
 
 Oxygen 
 absorbed in 
 1 hour. 
 
 Ratio of 
 oxygen in 
 3 hours to 
 oxygen in 
 1 hour. 
 
 hour. 
 
 3 hours. 
 
 Edinburgh City supply 
 
 0-76 
 0-91 
 0-86 
 0-04 
 
 o-oo 
 
 0-02 
 0-06 
 
 o-oo 
 
 0-56 
 0-06 
 0-26 
 0-70 
 0-92 
 0-56 
 
 o-oo 
 o-oo 
 
 0-94 
 0-88 
 1-23 
 1-26 
 0-80 
 2-43 
 2-29 
 0-87 
 0-37 
 
 2-18 
 1-93 
 1-66 
 0-13 
 
 o-oo 
 
 0-07 
 0-21 
 0-12 
 1-60 
 0-11 
 0-42 
 1-86 
 1-80 
 1-24 
 0-33 
 
 o-oo 
 
 1-96 
 1-43 
 2-70 
 2-98 
 1-57 
 7-16 
 6-92 
 1-71 
 0-75 
 
 5-94 
 6-30 
 5'32 
 0-99 
 0-20 
 0-91 
 1-35 
 1-43 
 4-20 
 1-05 
 2-25 
 8-66 
 7-42 
 4-96 
 5-69 
 
 o-oo 
 
 10-42 
 10-28 
 10-49 
 11-12 
 6-76 
 32-10 
 32-80 
 5-90 
 3-60 
 
 2-7 
 3-1 
 3-2 
 
 7-6 
 
 13-0 
 6-4 
 12-0 
 2-6 
 9-5 
 5-9 
 4-6 
 4-1 
 4-0 
 
 8-3 
 7-2 
 3-8 
 3-8 
 4-2 
 4-4 
 4-7 
 3-4 
 4-8 
 
 
 
 St Margaret's Well 
 
 St Anthony's Well 
 
 Spring, Braids 
 
 Blackford Hill 
 School Well Straiton 
 
 Stream 
 I. Well 
 
 II. 
 
 I. Prison, Peterhead 
 II. 
 III. Well 
 IV. 
 Spring, Edinburgh 
 
 **> fSt Margaret's Loch ... 
 
 5 ... 
 .2 Duddington ... 
 g | ... 
 jg .2 -| Dunsappy 
 'V Water of Leith ... 
 
 
 -g Stream Liberton 
 fa Braids 
 
 
of Potable Waters 
 
 51 
 
 TABLE XIX 
 
 Table showing that on boiling an Organic Solution con- 
 taining Sulphuric Acid and Permanganate, Oxida- 
 tion is very Rapid during the First Quarter of an 
 Hour and much more Gradual afterwards. 
 
 Oxygen is stated as parts per million. 
 
 Date. 
 
 Water. 
 
 Oxygen absorbed in 
 
 i hour. 
 
 | hour. 
 
 1 hour. 
 
 2 hours. 
 
 3 hours. 
 
 1888 
 
 
 
 
 
 
 
 Nov. 22 
 
 Edinburgh City supply 
 
 5-16 
 
 
 
 5-94 
 
 
 
 
 
 ,, 24 
 
 
 
 4-54 
 
 4-88 
 
 5-36 
 
 
 
 
 
 26 
 
 
 
 
 
 
 5-74 
 
 5-98 
 
 
 
 27 
 
 
 
 
 
 
 
 
 
 
 
 6'04 
 
 29 
 
 JJ 
 
 
 
 
 
 7-02 
 
 
 
 7'96 
 
 1889 
 
 
 
 
 
 
 
 Apr. 20 
 
 Dim Loch, Peaty 
 
 
 
 
 
 27-10 
 
 28-21 
 
 
 
 8 
 
 I. Well, Peterhead 
 
 
 
 
 
 7-56 
 
 9-12 
 
 
 
 )> 8 
 
 II. ,. 
 
 
 
 
 3-35 
 
 4-28 
 
 ~ 
 
52 The Organic Analysis 
 
 Comparison between the Results obtained by the Oxygen 
 Process at Ordinary Temperatures, and at 100 C. 
 
 I. Made-up (tyganic Solutions (see Table XVI). 
 At ordinary temperatures, cane sugar practically ab- 
 sorbed no oxygen from permanganate during the first 
 three hours, whereas, at 100 C. for one hour 10 milli- 
 grammes of cane sugar absorbed 9*74 milligrms. 
 oxygen, and at the end of two hours the sugar was 
 practically completely oxidised. 
 
 A solution of starch, when kept at 100 C. for two 
 hours, absorbs upwards of 400 times more oxygen than 
 during three hours in the cold. 
 
 Albuminoid solution from potato absorbed five times 
 as much oxygen, and from wheaten flour sixteen times 
 as much oxygen in the heat as during three hours in 
 the cold. 
 
 Urine absorbed five times as much oxygen at 100 
 C. for two hours as in the cold for three hours. 
 
 Vegetable alkaloids are practically completely oxi- 
 dised when kept at 100 C. for two hours, in some 
 cases at the end of one hour, whereas in the cold not 
 more than a quarter of the required oxygen is ab- 
 sorbed in three hours. 
 
 Conclusions from Table XVI. (1) Some organic 
 solutions, such as sugar and vegetable alkaloids, are 
 
of Potable Waters 53 
 
 completely oxidised if boiled for two hours with sul- 
 phuric acid and permanganate. 
 
 (2) Albumens are only partially oxidised by boiling 
 in this way for two hours, the ratio between oxygen 
 absorbed and oxygen required for complete oxidation 
 seeming to vary with the state the albuminoid body 
 is in. For example, the albumen from potato was 
 completely oxidised at the end of two hours, whereas 
 the solution of moist white of egg only absorbed 
 about seven-sixteenths of the oxygen required to oxi- 
 dise the albumen only of the moist white of egg. 
 
 (3) Starch, if boiled for two hours with acid and 
 permanganate, absorbs rather less than three quarters 
 of the oxygen required for complete oxidation. 
 
 II. Natural Waters (see Table XVII) .This table 
 shows that the oxygen absorbed by waters, when 
 boiled for two hours with sulphuric acid and perman- 
 ganate, exceeds the quantity absorbed during three 
 hours in the cold, in a ratio ranging from 2-5 to 1 up 
 to 8 to 1 in the case of potable waters and polluted 
 waters. 
 
 Oxidation of organic matter is very rapid as soon 
 as the organic solution is brought to the boil, and after 
 the first quarter of an hour the oxidation is much 
 more gradual. 
 
54 The Organic Analysis 
 
 Classification of Waters for the Oxygen Process at 
 100 G. for two Hours. 
 
 A water is of great purity if it absorbs less oxygen 
 than 2 parts per million. 
 
 It is of mediunt purity if the oxygen absorbed is 
 between 2 and 4 parts per million. 
 
 A water is suspicious, unless peaty, if the oxygen 
 absorbed is between 4 and 6 parts per million. 
 
 A water is not to be recommended, unless peaty, if 
 the oxygen absorbed is over 6 parts per million. 
 
 In any case, the oxygen absorbed should not be 
 much over 6 parts per million. 
 
of Potable Waters 55 
 
 PAET III 
 
 THE SULPHURIC ACID PROCESS FOR ESTIMATING 
 THE ORGANIC NITROGEN 
 
 THE chief objections to the Albuminoid Ammonia 
 Process are 
 
 (1) The ammonia obtained by this process does not 
 represent the whole of the nitrogen of albuminoid 
 bodies. Mr Wanklyn states that albumen yields about 
 two-thirds of its nitrogen as ammonia to alkaline per- 
 manganate, and the analyses contained in the follow- 
 ing tables bear out this statement. Albuminoid 
 bodies play an important part in the organic impuri- 
 ties of waters, being present whether the organic 
 matter is of animal or vegetable origin. 
 
 It will be shown later on that urine, even when 
 highly diluted, only yields half of its nitrogen as am- 
 monia to alkaline permanganate. 
 
 Vegetable alkaloids yield varying proportions of 
 their nitrogen as ammonia by Mr Wanklyn's process. 
 
56 The Organic Analysis 
 
 Although, the same water may yield a fairly constant 
 proportion of its nitrogen as ammonia to alkaline per- 
 manganate, when analysed from time to time, yet two 
 waters taken from different sources do not necessarily 
 yield up the nitrogen in quantities having the same 
 ratio to the organic matters contained in these waters. 
 This is due to the ratio of nitrogen yielded as ammonia 
 to total nitrogen present being different for different 
 organic substances, or different for the same organic 
 bodies existing under different conditions. For ex- 
 ample, the albuminoid solution from potato yielded up 
 about nine-tenths of its nitrogen as ammonia to alka- 
 line permanganate, whereas an albuminoid solution 
 from wheat-flour only yielded up about three quarters 
 of its nitrogen as ammonia. 
 
 (2) Occasionally, waters are met with, which yield 
 up the nitrogen as ammonia so slowly that it is diffi- 
 cult to exhaust the water of its albuminoid ammonia. 
 This point may be of no importance in determining 
 the fitness, or unfitness, of the water for domestic pur- 
 poses, but it imparts a degree of incompleteness to the 
 operation. 
 
 (3) Solutions of alkaline permanganate, even when 
 carefully prepared, contain traces of ammonia; and this 
 has to be allowed for in each analysis, or the solution 
 of alkaline permanganate must be boiled free of am- 
 monia, just previous to use. The caustic potash seems 
 to contain a trace of organic matter, which is gradu- 
 
of Potable Waters 57 
 
 ally acted upon by the alkaline permanganate, giving 
 rise to ammonia. 
 
 The process for estimating the organic nitrogen, 
 brought out by Drs Frankland and Armstrong, has 
 failed to satisfy analysts. Its alleged inaccuracy, 
 the delicacy of apparatus required, and the necessity of 
 gas-analysis, have prevented this process from coming 
 into general use. 
 
 Kjeldahl in 1883 published a paper proving that, if 
 organic substances containing nitrogen are charred by 
 strong sulphuric acid in the heat, all the nitrogen is 
 converted into sulphate of ammonia, unless the nitro- 
 gen is present as azo-, diazo-, or nitro-compounds. He 
 also proved that the results obtained are as accurate 
 as by the " Will and Yarrentrapp Process." Kjeldahl 
 applied the process to solid or semi- solid organic com- 
 pounds, and used oxalic acid to estimate the quantity 
 of ammonia obtained. 
 
 Is the Sulphuric Acid Process applicable to the Analysis 
 of the Organic Matters contained in Potable Waters ? 
 
 In order to answer this question the following points 
 must be considered : 
 
 I. During the evaporating down of a liquid con- 
 taining ammonia, is any of the ammonia lost if the 
 
58 The Organic Analysis 
 
 liquid be strongly acidified by sulphuric acid previous 
 to heating ? 
 
 II. If water containing both free and organic am- 
 monia is evaporated down to a small bulk say one fifth 
 of its original volume without adding any acid, will 
 the free ammonia only be driven off, or will some of 
 the organic ammonia be lost due to the heat decom- 
 posing part of the organic bodies present ? 
 
 In answer to the first question, it may be shown 
 that a solution of a known quantity of ammonium 
 chloride, when strongly acidified with pure sulphuric- 
 acid, can be subjected to the process of distillation 
 without the distillate containing any ammonia. It 
 may be also proved that the ammonia is in the resi- 
 dual water left in the retort after distillation, if the 
 solution be rendered alkaline by a solution of caustic; 
 soda, which has been boiled free of all ammonia, and 
 distillation resumed. It is found, that the distillate* 
 now contains all the ammonia originally added, as 
 well as any ammonia the sulphuric acid may have con- 
 tained. 
 
 The experiments with natural waters (see Table 
 XX) give the answer to both questions. In this table 
 it is shown that if a water is selected which contains 
 both free and organic ammonia ; if the free ammonia 
 is measured by taking J or 1 litre of the water and 
 distilling off all the free ammonia, and nesslerising 
 the distillate ; if the total ammonia free and organic 
 
of Potable Waters 
 
 is found by subjecting a litre of the water to the- 
 sulphuric acid process; if the organic ammonia is 
 found by boiling down a J litre of the water to 50 c.c.,. 
 then adding 10 c.c. of pure sulphuric acid and con- 
 ducting the sulphuric acid process to be immediately 
 described; then it is found that the total ammonia 
 obtained is equal to the sum of the free and organic 
 ammonia. 
 
 These experiments prove 
 
 (1) That an organic solution, when rendered strongly 
 acid by sulphuric acid, does not lose any of its con- 
 tained ammonia during the evaporation of the water, 
 and the charring by sulphuric acid. 
 
 (2) That a neutral organic solution can be evapo- 
 rated down to, at least, one-fifth of its original volume 
 without any loss of organic ammonia. 
 
 The waters experimented with in Table XX are of 
 various degrees of purity, ranging from the pure well 
 water from Patna down to the Water of Leith water 
 at Bonnington, which is little better than sewage 
 water. It should be borne in mind that the ammonia 
 was measured in the usual way by Nessler's reagent, 
 and that by this plan the experimental error may be- 
 ToVo" ^ a niilligrm. of ammonia for each 50 c.c. 
 of distillate nesslerised, and that when J litre of 
 water is used the total error must be multiplied by 
 four if the ammonia is stated as parts per million. 
 Bearing these facts in mind, it is seen how closely the 
 
60 The Organic Analysis 
 
 three operations for the free, organic, and total 
 ammonia tally with one another. It follows that the 
 sulphuric acid process for organic nitrogen is appli- 
 cable to potable waters. 
 
 Working Details for the Sulphuric Acid Process for 
 Organic Nitrogen. 
 
 Apparatus and reagents necessary : 
 
 (a) A glass-stoppered retort of f litre capacity. 
 
 (6) A distillation flask of J litre capacity. 
 
 (c) A Liebig's condenser and stand. Also an iron 
 stand with rings. 
 
 (d) A litre measuring flask ; a 10 c.c. and a 2 c.c. 
 pipette ; a 50 c.c. burette, with Geissler's stopcock, 
 graduated to -^ of a c.c. ; several nesslerising 
 glasses, each marked at 50 c.c. ; several glass-stop- 
 pered flasks of 50 c.c. capacity, with the 50 c.c. mark 
 in neck of flask ; also a 100 c.c. glass-stoppered flask. 
 
 (e) Nessler's reagent. 
 
 (/) A weak solution of ammonium chloride. 
 (g) Distilled water, 50 c.c. of which should give no 
 colour reaction with Nessler's reagent. 
 (h) A solution of pure arsenious acid. 
 For preparation of these reagents see Appendix A. 
 (i) Pure sulphuric acid. 
 & Solution of caustic soda. 
 
of Potable Waters 61 
 
 Preparation of Pure Sulphuric Acid. 
 
 The pure distilled acid can be bought, and, if 10 c.c, 
 of the acid do not contain more than 0*02 milligrm. 
 of ammonia, the acid may be used as it is, or subjected 
 to the following process : 
 
 If the distilled acid cannot be obtained, the acid 
 may be distilled in the laboratory from a glass-stop- 
 pered retort into an air-tight glass receiver. The 
 acid may be prepared very pure in this way, but the 
 process is slow; generally 10 c.c. of this distilled 
 acid will contain 0'005 milligrm. of ammonia. An 
 easier process for purifying the acid of its ammonia is 
 based upon the fact that nitrous acid and ammonia, 
 when heated together, mutually destroy one another, 
 forming water and free nitrogen, thus : NH 3 + HN0 2 
 = 2H 2 + N a . 
 
 Manufacturers of sulphuric acid, who use platinum 
 retorts, are in the habit of adding sulphate of ammonia 
 to the acid in order to destroy the nitrous acid in 
 this way, as it injures the platinum retorts. 
 
 Mr Raphael Meldola and Dr E. E. Moritz* have 
 drawn attention to the reverse process, namely, that 
 by adding excess of potassium nitrite to sulphuric 
 acid and heating for two hours and a quarter, the 
 
 * See number of ' Journal of Society of Chemical Industry ' for 
 February 29th, 1888. 
 
62 The Organic Analysis 
 
 ammonia is destroyed, unless the nitrogen is present 
 as a nitro-, nitroso-, azo-, or diazo-compound. 
 
 I find the best way to obtain the acid free of 
 ammonia is as follows : Find out by experiment how 
 much ammonia 5 c.c. of the sulphuric acid contain. 
 This is done by neutralising and making alkaline the 
 diluted acid with a solution of caustic soda boiled free 
 of ammonia, and nesslerising the distillate. Next 
 take 100 c.c. or J- litre of the sulphuric acid; put it 
 into a clean glass-stoppered retort ; add sufficient per- 
 manganate of potash to render the acid moderately 
 green in colour, and heat on a sand-bath till the acid 
 becomes again colourless. This ensures the complete 
 oxidation of any organic compounds which may have 
 got into the acid since it was manufactured. Calculate 
 the quantity of potassium nitrite necessary to destroy 
 all the ammonia contained in the acid, and add excess 
 of nitrite to the acid ; continue to heat the acid on a 
 sand-bath. Collect the distilled acid as it drops from 
 the end of the retort, and test, from time to time, for 
 nitrous acid by collecting a few drops of the distillate, 
 diluting it with water, and adding a drop of a solu- 
 tion of potassium iodide; add a few drops of starch 
 solution, and if nitrous acid is present the solution 
 strikes a blue colour. Continue to heat the sulphuric 
 acid till there is no blue reaction obtained in this way 
 from the distillate. This precaution ensures that the 
 acid in the retort is now free of any nitrous acid. 
 
of Potable Waters 63 
 
 Now take 10 c.c. of the acid in the retort ; drop it into 
 30 c.c. of ammonia-free distilled water ; pour it into 
 a clean distillation flask, and render the solution 
 alkaline with ammonia-free solution of caustic soda. 
 On distilling and nesslerising the distillate there is 
 generally no colour reaction obtained. If there is 
 any ammonia still present the quantity is very small , 
 and if the mean of three estimations of the ammonia 
 is taken, the quantity of ammonia is a perfectly con- 
 stant factor. The cold acid should now be run out of 
 the retort into a perfectly clean J litre bottle, having 
 a well-fitting glass stopper. 
 
 Preparation of the Solution of Caustic Soda. 
 
 The solution, when ready for use, should contain 
 neither ammonia nor nitrous acid. The commercial 
 caustic soda in stick sometimes contains a good 
 deal of nitrite. If a solution containing a small 
 quantity of sulphate of ammonia is rendered alkaline 
 with a solution of caustic soda containing sodium 
 nitrite, and submitted to distillation, there is a slight 
 loss of ammonia. 
 
 Caustic soda is better than caustic potash, as a solu- 
 tion of sodium sulphate can be concentrated further 
 than one of potassium sulphate before it begins to 
 bump. 
 
64 The Organic Analysis 
 
 If pure caustic soda cannot be obtained, a very pure 
 solution of it may be prepared in the following way. 
 Procure crystallised sodium carbonate, and if it con- 
 tains nitrous acid, free it of this impurity by recrystal- 
 lising the carbonate. In a litre of boiling water good 
 tap- water does if it is free of nitrates, nitrites, and con- 
 tains little organic matter dissolve 350 grms. of this 
 sodium carbonate. Bring the solution to the boil in a 
 large flask, add 75 grms. of slaked lime, freshly pre- 
 pared from properly burned limestone; continue to 
 boil the mixture on a sand-bath for ten minutes, then 
 remove and let stand till the supernatant liquid is quite 
 clear. The clear liquid is syphoned off into a glass- 
 stoppered bottle. A solution so prepared contains no 
 nitrites, very little ammonia, and practically no organic 
 matter. The strength of the caustic solution is about 
 8 per cent, of sodium hydrate, and a J litre of the solu- 
 tion will more than neutralise 10 c.c. of sulphuric acid. 
 This solution, or one of similar strength prepared from 
 caustic soda in stick, when about to be used, is boiled 
 in a distillation flask or retort till 25 c.c. of the distil- 
 late give no colour reaction with Nessler's reagent. 
 
 The Sulphuric Acid Process. 
 
 If the retort has not been used for the same purpose 
 before, it must be made clean by heating some sul- 
 
of Potable Waters 67 
 
 phuric acid in it for some time, so as to char off all 
 traces of organic matter. The retort is afterwards 
 well washed out with tap water till the drippings do 
 not redden blue litmus paper. Connect the retort 
 with a Liebig's condenser, in the usual way, support- 
 ing the retort on a piece of wire-gauze covering the 
 ring of an iron stand. The retort is fixed securely by 
 passing another ring of the iron stand over the tubu- 
 lated part of the retort. Measure out a J litre of the 
 sample of water, and pour it into the retort through a 
 glass funnel. Replace the stopper of the retort, and 
 adjust the condenser so that the inner third of the 
 funnel-shaped end of the retort is at an angle to allow 
 the water to run back into the retort ', this arrange- 
 ment prevents any loss of the water by bumping 
 during the boiling process. If the water is acid, add 
 a little freshly ignited sodium carbonate. Boil the 
 water briskly till its bulk is reduced to about 50 c.c., 
 then let it cool down. When nearly cold, add 10 c.c. 
 of the sulphuric acid, which has been prepared in the 
 way already described, also 10 c.c. of the solution of 
 sodium arsenite. Replace the wire-gauze by a small 
 sand-bath, and apply heat more gently. The remain- 
 ing 60 c.c. of water will gradually distil off, and the 
 distillate must be tested to see if it contain any nitrous 
 acid ; the most delicate test for nitrous acid is potas- 
 sium iodide and starch. If nitrates are present in the 
 water, the arsenious acid will reduce the nitric acid, 
 
68 The Organic Analysis 
 
 converting it into lower oxides of nitrogen, which 
 ultimately nearly all collect in the distillate as nitrous 
 acid. If the distillate contain no nitrous acid, the 
 water in the retort is allowed to distil off unin- 
 terruptedly till the sulphuric acid begins to fame. If 
 nitrous acid distils over, the water must not be allowed 
 to concentrate sufficiently to allow the sulphuric acid 
 to char, or some of the organic ammonia will be de- 
 stroyed by the nitrous acid. To prevent this over- 
 concentration, 30 c.c. of distilled water must be added 
 when the contents of the retort are reduced to about 
 30 c.c. if, when this stage is reached, there is still 
 nitrous acid distilling over. This addition of distilled 
 water may require to be repeated if there is much 
 nitric acid. Whenever a few drops of the distillate 
 give no colour reaction with potassium iodide and 
 starch, the concentration of the sulphuric acid is 
 allowed to go on. Hard waters are certain to bump 
 whilst the last 10 or 20 c.c. of water are distilling off, 
 and the flame must be moderated sufficiently to pre- 
 vent this bumping. If a few small pieces of platinum - 
 foil have been put into the retort the foil should be 
 heated to redness before it is put into the retort the 
 bumping is less likely to occur. Whenever the acid 
 is sufficiently concentrated to char, it will assume a 
 brownish colour, if there is much organic matter in 
 the water; this is shown well in peaty waters and 
 sewages, as the quantity of organic carbon is con- 
 
of Potable Waters 69 
 
 siderable. Purer waters will scarcely tinge the sul- 
 phuric acid. Whenever the acid is sufficiently strong 
 to fume freely the Liebig condenser is raised higher, 
 and this allows the acid fumes to condense back into 
 the retort from its funnel-shaped end. During the 
 charring process the flame may be increased, as the 
 liquid ceases to bump as soon as the acid begins to 
 fume. 
 
 Continue to heat the acid till it again becomes 
 colourless, and in cases where it has not been coloured 
 at all, it is well to keep the acid fuming for two hours. 
 Hence, in all cases, it is advisable to keep the acid 
 fuming for two hours at least, and then allow it to cool. 
 
 When the acid is cold, carefully drop in 30 c.c. of 
 ammonia-free distilled water, so as not to crack the 
 retort by overheating it. Now disconnect the retort 
 from the condenser, and flush the condenser well with 
 tap water, in order to wash out any sulphuric acid on 
 the inside tube. The acid contents of the retort may 
 be now transferred to a distillation flask, or retained in 
 the retort. If the former course is followed, the acid 
 contents of the retort are carefully run into the dis- 
 tillation flask through a clean glass funnel ; the retort 
 is washed out with a little distilled water, and lastly 
 several times with some of the cold solution of caustic 
 soda, and the washings are poured into the distilla- 
 tion flask. Continue to run into the flask the caustic 
 soda solution till, after gentle agitation, the contents 
 
70 The Organic Analysis 
 
 of the flask are decidedly alkaline to litmus. If the 
 retort is preferred, some of the solution of caustic soda 
 should be run into the retort through its funnel- 
 shaped end, so as to wash back into the retort the 
 acid adhering to the glass,, and more caustic solution 
 is added till the liquid in the retort is alkaline. Con- 
 nect the retort, or distillation flask, with the Liebig 
 condenser, using tinfoil as packing material if neces- 
 sary. Bring the contents of the retort to the boil, and 
 distil, at a moderate rate, into the 100 c.c. glass-stop- 
 pered flask. The necessary quantity of caustic soda 
 solution generally makes the contents of the retort up 
 to about 250 c.c., and by the time 100 c.c. have been 
 distilled off all the ammonia will be in the distillate, if 
 distillation has not been too rapid. 
 
 The Nesslerising Process. Have some clean nessler- 
 ising glasses ready standing on a white tile, and 
 some of the weak solution of ammonium chloride in a 
 burette. Shake up well the contents of the 100 c.c. 
 flask containing the distillate, and then divide the dis- 
 tillate between two of the 50 c.c. stoppered flasks. By 
 means of a pipette, remove 10 c.c. of the distillate 
 from one of the flasks, and pour it into a nesslerising 
 glass ; add 2 c.c. of Nessler's reagent ; if the ammonia 
 present seems to be less than -^ milligrm. ( = 2 c.c. 
 of the solution of ammonium chloride), rapidly add 
 the remaining contents of the flask from which the 
 10 c.c. were removed. Try to imitate the tint of 
 
of Potable Waters 71 
 
 colour obtained, by running into another nesslerising 
 glass as much solution of ammonium chloride as you 
 think necessary; make the volume up to 50 c.c. with 
 ammonia-free distilled water, and then add 2 c.c. of 
 Nessler's reagent. Gently mix the contents of the 
 glass; if the yellow tint obtained is too light, run 
 in more ammonium chloride solution till the shade of 
 colour is the same as that of the distillate. If the 
 shade of colour is too dark, a new portion must be 
 tried in another nesslerising glass. 
 
 The 50 c.c. of distillate in the other flask is used a& 
 a control experiment to prove the correctness of the 
 estimated quantity of ammonia in the first 50 c.c. If 
 10 c.c. of the distillate contain over ^ milligrm. of 
 ammonia, the quantity of ammonia in 10 or 20 c.c. of 
 the distillate should be measured, and the quantity in 
 100 c.c. of distillate thus calculated. When the ness- 
 lerising glass contains more than T ^- milligrm. of 
 ammonia, the difference in the shades of colour is not 
 so easily gauged as with a smaller quantity of ammonia. 
 It is important to have all the nesslerising glasses of 
 the same calibre, so that the 50 c.c. marks are all on 
 the same level, otherwise the same quantities of am- 
 monia may give different shades of colour. 
 
 Free Ammonia. If the condenser be washed out 
 before commencing distillation, the first 100 c.c. of the 
 200 c.c. of distillate taken off from the quarter of a 
 litre of water will contain all the free ammonia. Test if 
 
72 The Organic Analysis 
 
 there is much ammonia by nesslerising 10 c.c. of the 
 100 c.c. ; if the ammonia is small in quantity the 
 remaining 90 c.c. can be nesslerised in one glass, cor- 
 recting the quantity of ammonia obtained by allowing 
 for the 10 c.c. thrown away. If there is much ammonia 
 20 or 50 c.c. of the distillate can be standardised 
 against the solution of ammonium chloride. 
 
 The free ammonia and organic ammonia are ex- 
 pressed as parts per million, or milligrammes per litre, 
 if the respective quantities obtained from the quarter- 
 litre of water are multiplied by four. 
 
 The organic ammonia is expressed as organic nitro- 
 gen if the organic ammonia is multiplied by 0*824. 
 
 If the 10 c.c. of sulphuric acid used contained any 
 ammonia, and the water contained no nitrates or 
 nitrites, this quantity must be deducted from the 
 organic ammonia obtained, in order to find the correct 
 quantity of ammonia. If the 10 c.c. of acid only con- 
 tained a trace of ammonia say 0'005 to O'Ol of a 
 milligramme and the water contained nitrates, there 
 is no occasion to allow for this ammonia, as the nitrous 
 acid formed from the nitric acid will destroy this small 
 quantity of ammonia. 
 
 There is no occasion to add powdered permanganate 
 to the contents of the retort, during the charring pro- 
 cess, as the quantity of organic matter in a water is 
 so small that the sulphuric acid will completely oxi- 
 dise it within a reasonable time. There is always 
 
of Potable Waters 73 
 
 some danger of introducing ammonia, if powdered 
 permanganate is used. 
 
 Explanation of some of the Details of the Process. 
 
 The retort and Liebig's condenser are preferable 
 to small flat-bottomed flasks, because by their 'use 
 external ammonia is excluded, whereas with small 
 flasks ammonia is taken up from the gas flame and 
 fixed by the acid. 
 
 The Use of Arsenious Acid. This acid is superior to 
 ferrous chloride as a reducing agent in the sulphuric 
 .acid process, because an alkaline solution of arsenious 
 acid can be prepared free of ammonia, whereas fer- 
 rous chloride cannot be prepared in this way. 
 
 If a solution of ammonium chloride and potassium 
 nitrate is heated for two hours with sulphuric acid, 
 after all water is driven off, there is a partial loss of 
 ammonia, due to part of the nitric acid being reduced 
 to lower oxides of nitrogen and destroying some 
 ammonia. 
 
 If a water containing nitric acid and reducing 
 agents, such as ferrous salts, is operated upon by the 
 sulphuric acid process, there is a partial or complete 
 destruction of the organic ammonia, according to cir- 
 cumstances. 
 
 It is for these reasons that the arsenious acid is 
 
74 The Organic Analysis 
 
 added to the water when its bulk is reduced to 50 c.c., 
 in order to destroy all the nitric acid, if present, 
 before the sulphuric acid begins to char the organic 
 matter. 
 
 Remarks on the Sulphuric Acid Process. The expe- 
 riments contained in the various tables prove that by 
 this proc9ss it is possible to accurately measure the 
 total nitrogen contained in albuminoid bodies, and 
 other organic substances usually present in potable 
 waters. 
 
 The process is not difficult to carry out, and if care 
 be taken to always use clean apparatus and flush 
 pipettes, &c., with tap water just previous to use, the 
 sources of external ammonia are practically nil. It 
 is well to keep the retort for this process only, and 
 use it for nothing else. It is not advisable to distil, 
 or nesslerise, in an atmosphere containing much am- 
 monia. 
 
of Potable Waters 
 
 TABLE XX 
 
 Table showing that on boiling down a Potable, or Pol- 
 luted Water, to one-fifth of its Original Volume 
 practically only Free Ammonia is driven off, and 
 that no Organic Ammonia is lost. 
 
 Total ammonia obtained by adding sulphuric acid 
 before heating the water. 
 
 Free ammonia estimated by nesslerising the distil- 
 late from \ litre of the water. 
 
 Organic ammonia obtained by adding sulphuric 
 acid after boiling the water down to one-fifth of its 
 original volume. 
 
 The ammonia is stated as parts per million. 
 
 Water. 
 
 M-ee 
 ammonia. 
 
 By sulpli 
 pro 
 
 ,. 
 Organic 
 ammonia. 
 
 uric acid 
 jess. 
 
 L s 
 
 Total 
 
 jinimonin. 
 
 Free am- 
 monia plus 
 organic 
 ammonia. 
 
 Edinburgh vva'er .. 
 
 0-015 
 0-150 
 1-696 
 1-040 
 0-050 
 0-008 
 0-032 
 0-024 
 
 0-290 
 0-230 
 0-320 
 3-690 
 0-145 
 0-104 
 0-512 
 0-320 
 
 0-310 
 0-370 
 2-020 
 4-690 
 0-190 
 0-115 
 0-530 
 0-330 
 
 0-305 
 0-380 
 2-016 
 4-730 
 0-195 
 0-112 
 0-544 
 0-344 
 
 I. Well, Peterhead 
 
 II. 
 Water of Leith Water 
 I. Well, Patna 
 
 II. Kirkmicliael 
 III. Peterhead 
 
 IV. 
 
 
76 
 
 The Organic Analysis 
 
 TABLE XXI 
 
 Table showing the Quantities of Ammonia obtained 
 from Solutions of known Substances by the Alkaline 
 Permanganate and Sulphuric Acid Processes. 
 
 In the experiments with strychnia, brucia, and 
 muriate of morphia the quantity of solution used only 
 contained 1 milligrm. of the substance. 
 
 In the case of urea only T ^ milligrm. was used. 
 Picric acid, 2 or 3 milligrms. were used. 
 
 The ammonia is stated as parts per milligramme 
 of substance. 
 
 Substance. 
 
 Alk. permanganate 
 Ammonia. 
 
 Sulphuric acid. 
 Ammonia. 
 
 Total nitrogen in 
 substance stated as- 
 ammonia. 
 
 Strychnia 
 
 0-070 
 
 0-1131 ~ 
 
 
 
 
 0-094 }' 103 
 
 0-101 
 
 Brucia 
 
 0-058 
 
 0-090 1 
 
 
 
 
 0-080 } ' 085 
 
 0-086 
 
 Muriate of morphia. . . 
 Urea 
 
 0-035 
 0'04 
 
 0-046 
 0-56 
 
 0-045 
 0-56 
 
 Picric acid 
 
 
 0-114"] 
 
 
 
 __ 
 
 0'122 1 n , 
 
 
 
 
 
 
 o-o96 r 1107 
 
 0-111 J 
 
 0-222 
 
 
 
 
 
of Potable Waters 
 
 77 
 
 TABLE XXII 
 
 Showing the Quantities of Ammonia obtained from 
 Solutions of unknown Quantities of Albuminoid 
 Substances by the Alkaline Permanganate and Sul- 
 phuric Acid Processes. 
 
 The ammonia is stated as parts per 100. 
 
 Substance. 
 
 Ammonia by alk. 
 permang. 
 
 Ammonia by sul- 
 phuric acid. 
 
 Urine 
 
 0'99 
 
 1-90 
 
 I. Solution of albumen from potato 
 
 2-97 
 
 3-38 
 
 II. M 
 
 5-88 
 
 7'00 
 
 flour 
 
 12-37 
 
 15-901 
 1 
 
 > > 
 
 
 
 15-15 J- 15-73 
 
 
 
 
 
 16-15J 
 
 Solution of flour 
 
 1-20 
 
 1-84 
 
 Moist white of eo'g 
 
 1-29 
 
 1-901 
 
 
 
 1 1-90 
 
 1-90 I 
 
 
 
 
78 
 
 The Organic Analysis 
 
 TABLE XXIII 
 
 Showing the Quantities of Ammonia obtained from 
 Ordinary Waters by the Alkaline Permanganate and 
 Sulphuric Acid Processes. 
 
 The ammonia is stated as parts per million. 
 
 Water. 
 
 Ammonia by Ammonia by sulphuric 
 a)k. prrni. acid. 
 
 I. Edinburgh City supply ... 
 Well Noi'l. Peterhead 
 
 0-126 
 0-132 
 0-170 
 0-152 
 
 o-ioo 
 
 0-178 
 0-194 
 0-120 
 0-037 
 0037 
 0-040 
 0-060 
 0-035 
 0-025 
 0-060 
 0-116 
 0-400 
 0-390 
 0-128 
 0-062 
 0-005 
 0-308 
 0-085 
 
 0-188 = 0-155 n 
 0-296 = 0-245 
 [0-670 = 0-552 
 0-550 = 0-453 
 0-230 = 0-189 
 0-320-0-263 
 jO-512 = 0-422 
 10-320 = 0-263 
 0-070 = 0-057 
 10-044 = 0-036 
 '0-104 = 0-857 
 ; 0-050 = 0-041 
 ,0-145 = 0-119 
 0-050 = 0-041 
 0-328 = 0-270 
 J0-300 = 0-247 
 0500 = 0-412 
 ;0-510 = 0-420 
 0-160 = 0-132 
 '0-164-0-135 
 
 itrogen. 
 
 n 
 
 > 
 )> 
 
 
 
 
 
 5> 
 > 
 
 it 
 3> 
 
 J) 
 
 it 
 
 
 
 2nd sample 
 Well No. II. Peterhead 
 III. 
 
 iv. 
 
 V. 
 I. Kirkmichael 
 II. 
 HI- 
 
 iv. 
 
 I. Patna 
 
 II. ., 
 
 Water I. Eccles, Maidstone .. 
 
 ! ii.' " " '.'.'. 
 . ii. 
 , in. 
 
 ii. 
 
 '0-056-0-046 
 
 , in. 
 
 10-880-0-725 
 
 Water. Bearsted, Maidstone... 
 
 0-350 = 0-288 
 
of Potable Waters 
 
 79 
 
 TABLE XXIV 
 
 Showing the Quantities of Ammonia obtained from 
 Polluted Waters by the Alkaline Permanganate and 
 Sulphuric Acid Processes. 
 
 The ammonia is stated as parts per million. 
 
 Water. 
 
 Ammonia by 
 alk. perm. 
 
 Ammonia by sulphuric 
 acid. 
 
 I. Water of Leith, Bon- 
 nington 
 
 2'72 
 
 3-75 3'090 nitrogen 
 
 II. Ditto filtered through 
 clayey soil ... 
 
 2*38 
 
 3.89 3.205 
 
 Ill Sample II 
 
 2-06 
 
 2-29 1*186 
 
 IV. III 
 
 2-28 
 
 3-69-3-040 
 
 V. Ill and IV mixed after 
 precipitation by alum 
 VI. Duddingston Loch 
 VII. Dhu Loch, peaty water 
 VIII. Sewage water 
 
 0-80 
 0-28 
 0-29 
 99-00 
 
 0-98 = 0-807 
 0-99 = 0-815 
 0-72 = 0-593 
 190-00 156-560 , 
 
 
 
 
80 
 
 The Organic Analysis 
 
 TABLE XXV. 
 
 Showing the Quantities of Oxygen absorbed at the Ordi- 
 nary Temperature and at 100 0., also the Quantities 
 of Ammonia obtained by the Alkaline Permanganate 
 and Sulphuric Acid Processes. 
 
 Oxygen and ammonia are stated as parts per million. 
 
 
 Temp. 
 
 = ordinary. 
 
 Temp. 
 
 = 100C. 
 
 Ammonia 
 by 
 
 Ammonia by 
 
 Water. 
 
 Oxygen in 
 3 hours. 
 
 Oxygen in 
 2 hours. 
 
 alkaline 
 permang. 
 
 sulphuric acid. 
 
 Well I. Kirkmichael ... 
 II. 
 III. 
 IV. 
 I. Patna 
 II. 
 
 0-12 
 0-24 
 
 o-io 
 
 0-25 
 0-12 
 O'OO 
 
 0-90 
 0-60 
 0-88 
 1-54 
 0-73 
 0-50 
 
 0-037 
 0-037 
 0-040 
 0-060 
 0-035 
 0-025 
 
 Nitron- 
 0-044 = 0-036 
 0-070 = 0-057 
 0.104 = 0-857 
 0-050 = 0-041 
 0-145 = 0-119 
 0-050 0041 
 
 IV. Peterhead 
 V. 
 I Edinburgh water ... 
 II ... 
 Dhu Loch, peaty 
 
 2-24 
 
 0-86 
 1-62 
 1-46 
 975 
 
 9-12 
 4-28 
 6-13 
 5-72 
 28*21 
 
 0-194 
 0-120 
 0-126 
 0-132 
 
 0288 
 
 0-512 = 0-422 
 0-320=0-263 
 0-188 = 0-155 
 0-296 = 0-245 
 0720-0-593 
 
 
 
 
 
 
of Potable Waters 81 
 
 Remarks on the Tables dealing with the Sulphuric Acid 
 Process. 
 
 The analyses of known substances show that all 
 organic nitrogen is obtained as ammonia by this pro- 
 cess, whereas the nitrogen obtained by the alkaline 
 permanganate process fell short of the total ammonia, 
 in all cases. Picric acid is said not to yield any am- 
 monia to sulphuric acid, but in the sample analysed the 
 average quantity obtained represented about half of 
 the total nitrogen in the picric acid. It is an example 
 of a nitro-compound, and in these cases sulphuric acid 
 does not convert the nitrogen or only a part of the 
 nitrogen into ammonia. 
 
 Table XXII shows that the nitrogen of albuminoid 
 bodies is only partially converted into ammonia by 
 alkaline permanganate, the quantity obtained being 
 about two-thirds to three-fourths of the total nitrogen 
 present. The sulphuric acid process was used to esti- 
 mate the quantity of albuminoid bodies present. 
 
 Table XXV shows, at a glance, the very different 
 results obtained by the oxygen process at ordinary 
 temperatures and at 100 C., and by the alkaline per- 
 manganate and sulphuric acid processes. 
 
 6 
 
82 The Organic Analysis 
 
 A Classification of Waters based upon the Analyses of 
 the Waters in Tables XXIII and XXIV by the 
 Sulphuric Acid Process. 
 
 In waters of great organic purity the organic ni- 
 trogerl is less than O06 part per million. 
 
 In ordinary potable waters the organic nitrogen is 
 between O06 and O12 part per million. 
 
 Waters are suspicious (unless peaty) if the organic 
 nitrogen is over 0*12 part per million. 
 
 Peaty waters are potable if the organic nitrogen is 
 under O25 part per million. 
 
 Waters are condemned, even if peaty, if the organic 
 nitrogen is over 0*32 part per million. 
 
 It is not advisable to lay down hard and fast rules, 
 but these figures are approximately correct. 
 
of Potable Waters 83 
 
 PART IV 
 
 THE SULPHURIC ACID AND PERMANGANATE PROCESS 
 FOR ESTIMATING THE ORGANIC CARBON 
 
 WHEN a watery solution of carbo-hydrates is boiled 
 with sulphuric acid and permanganate there is complete, 
 or almost complete, oxidation of these substances into 
 carbonic acid and water. This is proved in the ex- 
 periments dealing with the oxygen process at 100 C. 
 In these experiments the quantity of oxygen required 
 to convert sugar into carbonic acid and water was 
 calculated, and it was found that the quantity of 
 oxygen absorbed by this substance tallied, or nearly 
 so, with the quantity calculated. It is evident that if 
 the concentration of the sulphuric acid solution is 
 pushed far enough the oxidising property of the acid 
 and permanganate will be greatly increased. Sulphuric 
 acid alone will oxidise carbon, if it is given time, bub 
 if aided by permanganate the oxidation process is very 
 rapid. Organic bodies when heated with sulphuric 
 
84 The Organic Analysis 
 
 acid are charred, some of the carbon separating out 
 as such, some passing off as carbonic dioxide, carbonic 
 monoxide, or other volatile carbon compound. If an 
 oxidising agent such as permanganate is present 
 whilst this charring is going on, the carbon passes off 
 as carbonic acid, and no carbonic monoxide escapes 
 this oxidising action. Oxalic acid is an illustration of 
 the conversion of carbonic monoxide into carbonic 
 dioxide when heated with sulphuric acid and perman- 
 ganate. If oxalic acid is heated with sulphuric acid 
 only, half of its carbon is converted into carbonic 
 monoxide, and half into carbonic dioxide. 
 
 The organic bodies present in ordinary waters are 
 completely disintegrated by sulphuric acid in the heat, 
 the nitrogen being converted into ammonia, the 
 hydrogen abstracted as water by the sulphuric acid, 
 and the carbon set free, or converted into volatile 
 carbon compounds. Under these circumstances, it is 
 a matter of experiment to find out if all the carbon of 
 these organic bodies can be obtained as carbonic acid, 
 when an oxidising agent is used along with the sul- 
 phuric acid. 
 
 In order to do this, two things are necessary, namely : 
 
 (1) Keagents containing no carbon compounds. 
 
 (2) Efficient means for collecting and estimating 
 the carbonic acid. 
 
of Potable Waters 85 
 
 Purity of Reagents. 
 
 Pure sulphuric acid, as it is sold, is generally free 
 of carbon as such, and of volatile carbon compounds. 
 To ensure that it contains no carbon compounds, 
 capable of oxidation into carbonic acid, it is only 
 necessary to heat some of the pure acid say 100 c.c. 
 with a little powdered permanganate of potash till 
 the acid again becomes colourless. 
 
 Permanganate of potash, when in solution, will in 
 time purify itself of traces of organic matter, especially 
 if the solution is acidulated with sulphuric acid. This 
 oxidation process is effected rapidly by boiling the 
 acidulated solution for some time. 
 
 The Collection and Estimation of the Carbonic Acid. 
 
 After trying various plans, all of which proved un- 
 satisfactory, I concluded that the apparatus used 
 ought to provide for 
 
 (1) The retention of the carbonic acid formed in 
 the apparatus containing the sulphuric acid, during 
 the charring process. 
 
 (2) The complete expulsion of the carbonic acid 
 from the apparatus, after the oxidation of the organic 
 substance is finished. 
 
86 The Organic Analysis of Potable Waters 
 
 (3) The exclusion of all extraneous carbonic acid 
 from the commencement to the finish of the process. 
 
 (4) The estimation of the carbonic acid formed. 
 The apparatus suitable for these purposes is as 
 
 follows : 
 
 (1) A retort (not tubulated) of 24 oz. capacity, 
 of moderately thick and uniform glass. 
 
 (2) A receiver of 6 oz. capacity, which fits accu- 
 rately to the retort, without allowing the retort to pro- 
 ject into the bulb of the receiver. 
 
 The retort and receiver are held firmly together, 
 when required, by a piece of stout india-rubber 
 tubing, overlapping the line of junction sufficiently to 
 render the joint air-tight to pressure from without. 
 
 The receiver is tubulated, and has fitted into the 
 hole a piece of glass tubing, secured properly by 
 stout rubber tubing overlapping the joint. 
 
 The piece of glass tubing is drawn out at its outer 
 end, so that | inch rubber tubing will pass over it, and 
 yet grasp the glass tubing firmly. Fit this end of 
 the glass tube with a piece of rubber tubing about 5 
 inches long. 
 
 For the collection of the carbonic acid formed, a small 
 flask of about 6 oz. capacity is required, fitted with 
 a rubber stopper through which passes a piece of 
 glass tubing, furnished at its outer end with a piece 
 of J inch rubber tubing about 5 inches in length. 
 
UFI7IRSITT 
 
 
 J 
 
The Organic Analysis of Potable Waters 89 
 
 Reagents. 
 
 The solutions required : 
 
 (1) Pure sulphuric acid. 
 
 (2) Solution of permanganate strength 10 grms, 
 to the litre. 
 
 (3) Solution of ferrous sulphate strength such 
 that 10 c.c. of the solution will decolorise 10 c.c. of 
 the permanganate solution. 
 
 (4) Centinormal solution of barium hydrate. 
 
 (5) Centinormal solution of oxalic acid. 
 
 (6) Solution of phenol-phthalein. 
 
 For preparation of these solutions see Appendix E. 
 
 Working Details of the Process. 
 
 The retort is charred free of organic matter unless 
 used before by heating some sulphuric acid in it. 
 When purified in this way, wash it out well with tap 
 water. Pour into the retort a litre of the sample 
 of water; support the retort by means of a retort 
 stand, and protect the bottom of the retort from the 
 gas flame by means of a square of wire-gauze cover- 
 ing the ring of an iron stand. Briskly boil the 
 water till its bulk is reduced to about 100 c.c., then 
 add 2 c.c. of pure sulphuric acid, having previously 
 diluted it ; continue to boil till the water is reduced 
 
DO The Organic Analysis 
 
 to about 50 c.c. All carbonates, likely to be in the 
 water, are decomposed on the addition of the acid, 
 and the carbonic acid is boiled off by the time the 
 volume is reduced to 50 c.c. At this stage, charge 
 the clean receiver with 10 c.c. of the permanganate 
 solution and 100 c.c. of distilled water; connect the 
 t receiver to the retort by folding the rubber tubing on 
 the retort halfway back, passing up the receiver to 
 the rubber tubing and then drawing the rubber 
 tubing down over the upper end of the receiver. If 
 the piece of rubber tubing is about 2 inches long, 
 and the connection properly made, this joint will be 
 air-tight to any pressure exerted during the process. 
 By means of a second Bunsen burner the perman- 
 ganate solution is brought to the boil ; the boiling of 
 the water residue has been continued meanwhile ; the 
 permanganate solution in the receiver and the water 
 residue are boiled for five minutes, ensuring a good 
 flow of steam from the exit tube of the receiver. The 
 boiling solutions fill the whole apparatus with steam, 
 driving out the air and all carbonic acid in the retort 
 and receiver. At the end of five minutes, suddenly 
 clip the exit tube from the receiver, and remove both 
 gas flames from under the apparatus. The clipping 
 of the rubber tubing, attached to the glass tubing 
 inserted into the receiver, may be accomplished by a 
 metallic clip, or by simply kinking the rubber tubing 
 down over the end of the glass tubing, and fixing it 
 
of Potable Waters 91 
 
 clown by an elastic ring. If the joints are properly 
 made, no air will enter the apparatus, but if it does so 
 the fact is betrayed by the hissing noise made by the 
 air as it leaks through the jointing. Whilst boiling 
 the permanganate solution, previous to clipping the 
 rubber tubing, it is important that no permanganate 
 should bump into the water residue in the retort. 
 This is prevented by fixing the apparatus in such a 
 position that the retort is on a higher level than the 
 receiver. Whenever the tubing is clipped, and the 
 flames removed, tilt the whole apparatus so that the 
 permanganate solution can flow gently into the retort. 
 When the apparatus is in this position, it is easily 
 noted if bubbles of air are oozing through the joint 
 connecting the retort and receiver, as the air ascends 
 through the permanganate in a stream of small bubbles. 
 Previous to the addition of the permanganate solution, 
 the volume of the water residue will be about 25 to 
 30 c.c., and when about 30 c.c. of permanganate solu- 
 tion have flowed into the retort do not allow any more 
 to run in, but replace the apparatus in its former posi- 
 tion, so that the receiver is once more on a lower level 
 than the retort. Replace the wire-gauze by a small 
 sand-bath small concave plate of sheet iron covered 
 with sand and heat the contents of the retort very 
 gently. In about an hour the 50 c.c. of watery va- 
 pour will have distilled into the receiver, leaving the 
 acid and permanganate to act upon any organic matter 
 
92 The Organic Analysis 
 
 still unoxidised. After the sulphuric acid has fuuied 
 for about half an hour all permanganate will have 
 been converted into manganous sulphate, and the acid 
 will be colourless. If it has a yellowish tinge it indi- 
 cates carbon still unoxidised unless due to ferric 
 sulphate ; and when the acid has cooled another 10 
 c.c. of the permanganate solution must be allowed to 
 flow into the retort. 
 
 When the sulphuric acid is colourless all carbon is 
 now present as carbonic acid. 
 
 At this stage, the probable gaseous contents of the 
 retort are carbonic acid, oxygen, chlorine (iodine or 
 bromine rarely), and nitric and sulphuric acid fumes. 
 
 When the acid is cold run gradually into the retort 
 the residual permanganate solution ; agitate the con- 
 tents of the retort, allowing the solution to alternately 
 flow from receiver to retort, and from retort to receiver. 
 
 The total liquid contents of the retort are about 
 150 c.c. ; and on boiling these 150 c.c. what gases or 
 vapours will distil over ? The small quantity of nitric 
 acid present, in any water, is too highly diluted by 
 150 c.c. of water to distil over, as a nitric acid solu- 
 tion must have attained a strength of about 8 per cent, 
 before acid vapours pass off. The distillate will then 
 contain only carbonic acid, oxygen, chlorine (from the 
 chlorides in the water), and watery vapour. Only the 
 carbonic acid and chlorine require to be taken into 
 consideration. 
 
of Potable Waters 93 
 
 It is evident that if these two gases are distilled 
 into a solution of barium hydrate both will unite with 
 the baryta. The chlorine must be got rid of, as it will 
 interfere with the estimation of the carbonic acid ab- 
 sorbed by the barium hydrate. 
 
 For this purpose, the contents of the apparatus are 
 allowed to flow into the receiver, and then 10 c.c. of the 
 ferrous sulphate solution are sucked up into a 10 c.c. 
 pipette and passed into the receiver through the rubber 
 tubing on the exit tube of the receiver, taking care that 
 no air is allowed to enter into the receiver during this 
 operation. This is secured by fixing the rubber tubing 
 firmly on the end of the pipette, then unclipping the 
 tubing and allowing the solution to flow in slowly, and 
 by clipping the tubing again before the pipette is 
 completely emptied of its contents. Ten c.c. of the 
 ferrous sulphate solution will decolorise the perman- 
 ganate and still remain in excess, as some of the latter 
 has been destroyed already during the charring pro- 
 cess. The contents of the receiver are now very 
 gently heated oscillating the liquid and the gas 
 flame the whole time as the liquid is being heated 
 under diminished pressure, and will probably bump 
 very violently unless this precaution is taken. When 
 the liquid is heated to nearly the boiling-point, the 
 whole apparatus is detached from its fixtures and 
 taken into the hands, and the liquid allowed to flow 
 backwards and forwards between the receiver and 
 
94 The Organic Analysis 
 
 retort, shaking the contents well from time to time. 
 This operation ensures that all the chlorine is absorbed 
 by the liquid and acts on the ferrous sulphate, pro- 
 ducing ferric salt and hydrochloric acid. At the end 
 of ten minutes the reaction may be considered com- 
 pleted, and all chlorine converted into hydrochloric 
 t acid or ferrous chloride. 
 
 It only remains to boil off the carbonic acid into a 
 solution of barium hydrate, and estimate the carbonic 
 acid. 
 
 For this purpose the 6 oz. flask, fitted with the 
 rubber tubing, is partly filled with distilled water 
 the bottom of the flask should be covered with about 
 half an inch of water and the water boiled till all air 
 is expelled from the flask ; then clip the rubber tubing 
 in the usual way. When the flask has cooled run in 
 10 c.c. of the solution of barium hydrate, by means of 
 a pipette, in the same manner as when the ferrous 
 sulphate was passed into the receiver. It is essential 
 to get the whole of the baryta solution into the flask, 
 so the pipette must be allowed to empty itself till 
 only a few drops of the solution are left in it ; then clip 
 the tubing and quickly withdraw the pipette from the 
 tubing, when the remaining few drops of the solution 
 will be sucked out of the pipette into the rubber 
 tubing. Clip the end of the tubing with two fingers 
 and loosen the clip, allowing the remaining baryta 
 solution to flow into the flask. Restore the metallic 
 
of Potable Waters 95 
 
 clip to its former position, so as to prevent any air 
 getting in. Distribute the contents of the retort so 
 that half of the liquid is in the retort and half in the 
 receiver; attach the small flask to the receiver by 
 means of a connecting piece of glass tubing, with 
 .a bulb on it. The whole apparatus should be 
 adjusted so as to rest properly on its supports, and 
 the small flask is placed in a basin of cold water. 
 Unclip the rubber tubing next the receiver, and pro- 
 ceed to boil the liquid in the retort. To prevent vio- 
 lent bumping this must be done gently, agitating the 
 contents meanwhile, and using wire-gauze instead of 
 the sand-bath. A few small pieces of platinum-foil, 
 placed in the retort at the very beginning of the 
 analytical process, will moderate this tendency to 
 bumping. When the contents of the retort are boil- 
 ing leave them to boil, and begin to boil the contents 
 of the receiver in the same way ; when nearly boiling 
 unclip the tubing next the small flask, and adjust the 
 apparatus so that there are no kinks in the rubber 
 tubing. 
 
 The contents of the retort and receiver must be 
 kept boiling briskly for about seven minutes, and the 
 small flask kept cool by means of cold water. If the 
 operation has been performed properly, there should 
 be so little air in the apparatus that the rubber tubing 
 next the small flask always remains partly collapsed. 
 If the tubing becomes tense and hard some air 
 
The Organic Analysis 
 
 must have got in. The collapsed tubing shows that 
 there is no danger of any explosion, and it should be 
 the guide to indicate the state of pressure inside the 
 apparatus. After boiling for seven minutes, clip the 
 tubing attached to the small flask, and detach the 
 flask from its connections. 
 
 Briskly agitate the contents of the flask for five 
 minutes, and if the solution remains decidedly pink in 
 colour, outside air may be allowed to enter the flask 
 it is best to allow the purest air which can be got 
 to enter the flask, such as air out of doors. If the 
 solution is faintly pink or colourless another 10 c.c. 
 of the baryta solution must be run in, and the contents 
 of the flask again well shaken. 
 
 From a graduated burette the centinormal solu- 
 tion of oxalic acid is run into the flask, till the pink 
 colour just disappears on the addition of one "more 
 drop. Bead off the number of c.c. of oxalic acid 
 solution required to neutralise the contents of the 
 flask. 
 
 Next standardise the baryta solution by taking 10 
 c.c. of the solution, and finding how much oxalic acid 
 solution is required to neutralise the 10 c.c. The 
 mean of three experiments should be taken as the 
 proper strength of the baryta solution. If the number 
 of c.c. of oxalic acid solution used by the contents of 
 the flask is deducted from the number of c.c. used 
 by 10 c.c. of baryta solution, the difference represents 
 
of Potable Waters 97 
 
 the quantity of carbonic acid from the sample of 
 water. 
 
 For example, 10 c.c. of baryta solution required 
 9*6 c.c. of oxalic acid solution, and 10 c.c. of baryta 
 solution, plus the carbonic acid from a J litre of 
 water, used 1'5 c.c. of oxalic acid solution, therefore 
 the difference between these two numbers is 8*1 c.c. 
 of oxalic acid solution. 
 
 But 10 c.c. oxalic acid solution are equivalent to 
 4*40 milligrms. of carbonic acid, therefore 8'1 c.c. 
 
 of oxalic acid = ^ milligrms. of carbonic acid 
 
 = 3*564 milligrms. 
 Again, 4*4 milligrms. of carbonic acid = 1*2 milligrms. 
 
 of carbon, therefore - .. - = 0'972 milligrm. of 
 
 carbon. 
 
 When the answer is to be stated as carbon, the 
 equation is more simple, thus : 
 
 10 c.c. oxalic acid solution = 1*2 milligrms. carbon, 
 
 therefore ^- = carbon in the water stated as 
 
 milligrammes. 
 
 If much carbonic acid is formed from the organic 
 matter in the water there is a good deal of barium 
 carbonate formed, and the solution, when neutral- 
 ised with oxalic acid, will become pink again on 
 standing. This is due to the fact that barium car- 
 
 7 
 
98 The Organic Analysis 
 
 bonate is alkaline in reaction, so the quantity of 
 oxalic acid solution used at first to decolorise the 
 solution must be taken as the proper quantity. 
 
 I have not had yet an extended experience with 
 the application of this process for organic carbon to 
 potable waters, but the results so far have proved 
 satisfactory. 
 
 The solutions used to test the accuracy of the pro- 
 cess were carefully prepared, and the results obtained 
 show the method to be extremely accurate. 
 
 Very small quantities of the standard organic solu- 
 tions were purposely used, as the object in view was 
 the application of the process to water analysis. It 
 is evident that by using decinormal or normal solu- 
 tions of barium hydrate and oxalic acid, one would be 
 enabled to use much larger quantities of organic sub- 
 stances, without impairing the accuracy of the process. 
 
of Potable Waters 
 
 99 
 
 TABLE XXVI 
 
 Table giving the Results obtained when the Organic 
 Carbon of Standard Solutions was estimated by the 
 Sulphuric Acid and Permanganate Process. 
 
 All the solutions were made of the same strength, 
 so that 1 c.c. of the solution contained 1 milligrm. of 
 the pure organic substance. 
 
 Tartaric Acid. Kecrystallised and carefully dried. 
 
 Gallic Acid. On drying it was found to contain 
 9*5 per cent, of water of hydration. 
 
 Salicine. The pure substance used. 
 
 Sulphate of Quinine was carefully dried before use. 
 
 Substance. 
 
 Theoretical 
 
 carbon. 
 
 Carbon obtained. 
 
 Tartaric acid, 
 
 2 milligrms. used 
 
 0-640 
 
 0-678 "I Average for 2 milli- 
 
 )> 
 
 2 
 
 0-640 
 
 0-624 y grammes, 
 
 
 
 3 
 
 0-960 
 
 0-954 J 0-646. 
 
 Gallic 
 
 2 
 
 0-894 
 
 0-864 ] 
 
 
 
 2 
 
 0-894 
 
 0-924 [ 0-884. 
 
 
 
 2 
 
 0-894 
 
 0-864 J 
 
 Salicine 
 
 2 
 2 
 
 1-091 
 1-091 
 
 J96l Average for 2 mffli- 
 
 
 j> 
 
 2 
 H 
 
 1-091 
 0-818 
 
 1-104 f g ramm es, 
 
 -i.rjQO 
 
 0-876 J 1 090> 
 
 Quin. sulph. 
 
 1 
 
 0-643 
 
 0-642^1 
 
 
 
 H 
 
 0-964 
 
 0-804 Average for 1| 
 
 y> 
 
 1* 
 
 0-964 
 
 0'912 \- milligrammes, 
 
 )> 
 
 u 
 
 0-964 
 
 0-972 0-915. 
 
 > 
 
 1* 
 
 0-964 
 
 0-924J 
 
100 The Organic Analysis of Potable Waters 
 
 TABLE XXYII 
 
 Showing the Organic Carbon obtained from Natural 
 Waters by the Sulphuric Acid and Permanganate 
 Process. 
 
 Organic carbon is stated as parts per million. 
 
 Organic carbon per litre. 
 
 Water Company, Maidstone 0'72 
 
 0-60 
 
 Well, near Maidstone, alleged to be polluted ... 4'95 
 
APPENDICES 
 
The Organic Analysis of Potable Waters 103 
 
 APPENDIX A 
 
 Preparation of Solutions used in the Albuminoid 
 Ammonia Process. 
 
 (1) Nessler's Reagent. Dissolve 35 grms. of potas- 
 sium iodide and 13 grms. of corrosive sublimate 
 in 800 c.c. of distilled water, which has just been 
 heated to boiling ; then add successive small quanti- 
 ties of a cold saturated solution of corrosive sublimate 
 till the red iodide of mercury begins to be no longer 
 dissolved, but remains as a permanent precipitate. 
 Allow the solution to cool ; then render it strongly 
 alkaline by adding 160 grms. of solid caustic potash, 
 or 120 grms. of solid caustic soda, and make the 
 solution up to one litre with distilled water. When 
 the solution is cleared by settling, decant off the clear 
 liquid and render the reagent sensitive by adding 
 cautiously more cold saturated solution of corrosive 
 sublimate till a permanent dirty yellow precipitate 
 forms. Allow the reagent to settle in a well-stop- 
 pered bottle. When the liquid has again become clear 
 some of the reagent is decanted into a smaller bottle 
 
104 The Organic Analysis 
 
 for use. Properly prepared Nessler's reagent has a 
 yellowish tint, and at once strikes a yellow-brown 
 colour with 5 c.c. of the weak ammonium chloride 
 solution in 50 c.c. of water. 
 
 (2) Stock Solution of Ammonium Chloride. Dissolve 
 3' 15 grms. of pure ammonium chloride in crystals in 
 one litre of distilled water. Each cubic centimetre 
 of the solution contains 1 milligrm. of ammonia. 
 
 (3) The Weak Solution of Ammonium Chloride. 
 Take 10 c.c. of the above stock solution and make it 
 up to one litre with distilled water. Each cubic cen- 
 timetre of the solution contains y^- milligrm. of am- 
 monia. 
 
 This is the solution used for nesslerising. 
 
 (4) Solution of Alkaline Permanganate. Dissolve 
 200 grms. of solid caustic potash and 8 grms. of 
 pure permanganate of potash in one litre of dis- 
 tilled water ; boil in an open flask till the volume is 
 reduced to about 750 c.c., and make up to one litre 
 with ammonia-free distilled water. 
 
 Fifty c.c. of this solution are used for half a litre of 
 water in each analysis. 
 
 (5) Carbonate of Soda. Solution made by boiling, 
 for a short time, excess of the carbonate in water. 
 
 Ten c.c. of the solution are used, if the water to be 
 analysed is acid in reaction. 
 
 (6) Distilled Water. It must be free of ammonia. 
 This is easily obtained by distilling an ordinary potable 
 
of Potable Waters 105 
 
 water, which has been made slightly acid with a few 
 drops of sulphuric or phosphoric acid. The acid pre- 
 vents any ammonia from distilling over. A distilla- 
 tion flask or a still may be used. 
 
 These solutions (except No. 4) are also used for the 
 sulphuric acid process. 
 
 (7) Solution of Arsenious Acid. Used in the sul- 
 phuric acid process. 
 
 Dissolve by the aid of heat 50 grms. of the purest 
 arsenious oxide in one litre of distilled water, which has 
 been made moderately alkaline with caustic soda solu- 
 tion. The caustic soda solution must be free of 
 nitrates and nitrites (see the Sulphuric Acid Process) . 
 When the arsenious acid is all dissolved the solution 
 must be decidedly alkaline, or made so by adding 
 more caustic soda solution. Boil down in a flask 
 'till the solution is reduced from 1000 c.c. to 900 c.c., 
 and then make up to one litre with ammonia- free dis- 
 tilled water. 
 
 Ten c.c. of this solution are used for each operation. 
 
106 The Organic Analysis 
 
 APPENDIX B 
 
 Preparation of Solutions used in the Oxygen Process at 
 Ordinary Temperatures. 
 
 (1) Solution of Potassium Permanganate. Dissolve 
 0*395 grm. of the pure crystals in one litre of dis- 
 tilled water. 
 
 1 c.c. of this solution, with acid, yields 0*10 milli- 
 grm. of oxygen. 
 
 1 c.c. of this solution exactly oxidises 0*2875 milli- 
 grm. nitrous acid (N0 2 ). 
 
 1 c.c. of this solution exactly oxidises 0*2125 milli- 
 grm. hydrogen sulphide (H 2 S). 
 
 1 c.c. of this solution exactly oxidises 0*9000 milli- 
 grm. ferrous oxide (FeO). 
 
 (2) Dilute Sulphuric Acid. Add gradually 100 c.c. 
 of the pure acid to 300 c.c. of distilled water. The 
 basin containing the water should stand in cold water 
 whilst adding the acid, in order to prevent a too rapid 
 rise of temperature. 
 
 10 c.c. of this dilute acid are added to each sample 
 of water examined. 
 
 (3) Solution of Potassium Iodide. Dissolve 10 
 grms. of the purest iodide in 100 c.c. of distilled 
 
of Potable Waters 107 
 
 water ; five drops of this solution will decolorise more 
 than 10 c.c. of the permanganate of potash solution. 
 
 (4) Solution of Sodium Hyposulphite. Dissolve 
 0*775 grm. of the pure material in one litre of dis- 
 tilled water; about 40 c.c. of this solution will de- 
 colorise the iodine set free by 10 c.c. of the perman- 
 ganate solution. 
 
 (5) Solution of Starch. Add 1 grm. of powdered 
 starch to 250 c.c. of boiling distilled water, boil 
 for five minutes and filter. When the solution has 
 settled syphon off the clear liquid into a stoppered 
 bottle. A few drops of this solution will colour the 
 quarter-litre of water of a deep blue if there is free 
 iodine present. The solution gradually loses the 
 power of producing this blue colour with iodine. 
 
 The Chemical Reactions which occur at the various 
 stages of the Oxygen Process. 
 
 (1) Permanganate of potash in the presence of sul- 
 phuric acid is reduced by organic matter to potassium 
 sulphate and manganous sulphate. 
 
 K 2 Mn 2 8 + 3H 2 S0 4 + organic matter = K 2 S0 4 
 + 2MnS0 4 +3H 2 + 5 united with organic matter. 
 
 (2) When potassium iodide is added to a solution 
 of permanganate of potash in the presence of sulphuric 
 
108 The Organic Analysis 
 
 acid, potassium and manganous sulphates are formed 
 and iodine is liberated. 
 
 K 2 Mn 2 8 + 8H 2 S0 4 + 10KI = 6K 2 S0 4 + 2MnS0 4 
 + 8H 2 + 5I 2 . 
 
 (3) The addition of sodium hyposulphite to the 
 iodine solution decolorises the solution by producing 
 sodium iodide and tetrathionate of soda. 
 
 2Na 2 S 2 3 + I 2 = 2NaI + Na 2 S 4 6 . 
 
 (Sodium (Sodium (Sodium 
 
 hyposulphite.) iodide.) tetrathionate.) 
 
 (4) A solution of permanganate of potash when 
 boiled with hydrochloric acid is reduced to potassium 
 and manganous chlorides with the liberation of free 
 chlorine. 
 
 K 2 Mn 2 8 + 16HC1= 2KC1 + 2MnCl 2 + 8H 2 + 5C1 2 . 
 
 A water containing much chlorides when boiled 
 with permanganate and sulphuric acid will have its 
 chlorides partially decomposed, and chlorine set free. 
 The following reaction may be supposed to take place : 
 
 K 2 Mn 2 8 + 8H 2 S0 4 + 10NaCl= K 2 S0 4 + 2MnS0 4 
 
of Potable Waters 109 
 
 APPENDIX C 
 
 Preparation of the Solutions used in the Oxygen 
 Process at 100 C. 
 
 (1) Solution of Permanganate of Potash. Dissolve 
 1*975 grms. of pure crystals in 1 litre of distilled 
 water. 
 
 10 c.c. of this solution can yield 5 milligrms. of 
 oxygen. 
 
 (2) Dilute Sulphuric Acid. Add 100 c.c. of pure 
 acid to 900 c.c. of distilled water. 
 
 10 c.c. of this acid are used for each operation. 
 
 (3) Solution of Potassium Iodide. Use a 10 per 
 cent, solution of the pure material. 
 
 Twenty drops will decolorise more than 10 c.c. of 
 the permanganate solution. 
 
 (4) Solution of Sodium Hyposulphite. Dissolve 
 3*875 grms. of the pure material in 1 litre of dis- 
 tilled water. About 40 c.c. of this solution will de- 
 colorise the iodine set free by 10 c.c. of the perman- 
 ganate solution. 
 
 (5) Solution of Starch. Add 1 grm. of powdered 
 starch to 250 c.c. of boiling distilled water; boil 
 for five minutes and filter ; let stand till settled, and 
 syphon off the clear liquid into a stoppered bottle. 
 
110 The Organic Analysis 
 
 (6) Distilled Water. It is prepared free of organic 
 matter from an ordinary potable water by adding ex- 
 cess of permanganate of potash, so that the colour of 
 the water remains pink. Distil the water from a distil- 
 lation flask or still through a condenser into a flask 
 which fits closely to the condenser. Do not distil 
 too far, or some permanganate may get into the dis- 
 tillate. If the contents of the distillation flask are 
 made acid with sulphuric or phosphoric acid, the dis- 
 tillate will contain no ammonia. 
 
 APPENDIX D 
 
 The Estimation of Nitrates as Nitrous Acid by means 
 of Sulphuric Acid and Arsenious Oxide. 
 
 Whilst carrying out experiments with arsenious 
 acid in connection with the sulphuric acid process, 
 the results obtained led me to believe that, under cer- 
 tain conditions, the whole of the nitric acid present 
 in a sample of water could be collected in the distil- 
 late in the form of nitrous acid. 
 
 The conditions favorable to the total conversion 
 of nitric into nitrous acid by sulphuric acid and arse- 
 nious oxide are 
 
of Potable Waters 111 
 
 (1) The acid solution must not be concentrated too 
 much by distillation. 
 
 (2) The presence of air. 
 
 (3) Slow distillation. 
 
 (4) The quantity of nitric acid in the water must 
 not be too large. 
 
 In the sulphuric acid process the first condition is 
 present, as the concentration of the acid solution is not 
 carried beyond a reduction of volume to 30 c.c., till 
 all the nitric acid is reduced to lower oxides of nitro- 
 gen. This reducing action does not begin till the 
 sulphuric acid has attained a strength of 15 to 20 per 
 cent. 
 
 The second and third conditions are ensured by 
 distilling very slowly, and condensing the watery 
 vapour in a Liebig's condenser. 
 
 The fourth condition requires that the quantity of 
 water used should not contain more than 6 milligrms. 
 of nitric acid. If the nitric acid contained in the 
 i litre of water used for the sulphuric acid process 
 is estimated by this method, as well as the organic 
 nitrogen, then this is the outside quantity of nitric 
 acid the water should contain. 6 milligrms. of nitric 
 acid in J litre of water are equivalent to 1*68 grs. of 
 nitric acid per gallon. 
 
 The nitric acid is reduced to lower oxides of nitro- 
 gen, but in the presence of air the distillate ultimately 
 contains all the nitrogen as nitrous acid. If any 
 
112 The Organic Analysis 
 
 nitric oxide (NO) is formed, it takes up oxygen from 
 the air and becomes nitrous acid (HN0 2 ) . 
 
 If the nitric acid is large in quantity the nitric 
 oxide is probably formed too freely to become 
 properly oxidised, and some of it gets lost. 
 
 The process for collecting and estimating the 
 nitrous acid formed is as follows : 
 
 During the sulphuric acid process, arsenious acid 
 (=0*5 grm.) and sulphuric acid (=10 c.c.) are added 
 to the J litre of water when about 200 c.c. have been 
 distilled off. At this stage put into a small flask, 
 say of about 6 oz. in capacity, 1 c.c. of a caustic soda 
 solution which contains no nitrites, add 10 c.c. of 
 distilled water, and pass the lower end of the Liebig 
 condenser into the neck of the flask so that the dis- 
 tillate drops into the solution. The condenser should 
 be sloped at an angle, so that a drop of distillate 
 always blocks the lower end of the inside tube of the 
 condenser, and the upper end of the condenser should 
 be properly attached to the retort, so that there is 
 little chance of any gases escaping at this end of the 
 condenser. Continue to distil, adding distilled water 
 to the retort if necessary (see Sulphuric Acid Process), 
 till a drop of the distillate gives no immediate blue 
 reaction when acidulated with sulphuric acid, and a 
 little potassium iodide and starch added. 
 
 Wash out the condenser with a little distilled water, 
 and add the washings to the contents of the flask. 
 
of Potable Waters 113 
 
 The small flask now contains all the nitric acid as 
 sodium nitrite. It now remains to estimate the 
 nitrous acid in the distillate. This is done by means 
 of a standard solution of potassium nitrite, and a solu- 
 tion of metaphenylene-diamine. 
 
 The solution of metaphenylene-diamine is prepared 
 by dissolving O50 grm. of the substance in 200 c.c. 
 of distilled water, and adding sufficient sulphuric acid 
 to render the solution acid. 
 
 The solution of potassium nitrite is prepared by 
 dissolving 0-270 grm. of the purest material (or 
 equivalent quantities of silver nitrite and potassium 
 chloride) in one litre of water. Each c.c. of this solu- 
 tion is equivalent to 0'20 milligrm. of nitric acid. 
 Prepare a 1 in 4 solution of sulphuric acid. 
 
 Dilute the distillate up to 250 c.c. with water and 
 mix thoroughly. Measure out into a nesslerising glass 
 50 c.c. of the diluted distillate, add 2 c.c. of the dilute 
 sulphuric acid and 2 c.c. of the solution of metapheny- 
 lene-diamine. Agitate gently the contents of the 
 glass, and in a short time a yellow or reddish-yellow 
 colour will appear, according to the quantity of nitrous 
 acid present. The tint of colour must be imitated by 
 running some of the standard solution of nitrite from 
 a burette into another glass, diluting up to 50 c.c. 
 with water and adding 2 c.c. of each of the above 
 reagents. When an idea of the quantity of nitrous 
 acid present is obtained, take a fresh 50 c.c. of the 
 
 8 
 
114 The Organic Analysis 
 
 distillate, and the proper quantity of standard nitrite 
 solution, and proceed as above. It is best to start the 
 reaction in the solutions in the two glasses as near 
 the same time as possible, and in about ten minutes 
 the maximum tint of colour is reached. Fresh solu- 
 tions must be made up till similar tints are obtained 
 in both glasses. Fifty c.c. are the one fifth of the 
 distillate, and one c.c. of the nitrite solution is equi- 
 valent to milligrm. of nitric acid; therefore the 
 number of c.c. of nitrite solution which is equiva- 
 lent to 50 c.c. of the diluted distillate represents so 
 many milligrammes of nitric acid in the total distillate. 
 If the number of c.c. used of the standard nitrite 
 solution is multiplied by O222 the quantity of nitro- 
 gen (as HN0 3 ) in the water used is obtained. 
 
 If it is found that the nitric acid in the quarter-litre 
 of water is above 6 milligrms., a smaller quantity 
 of water must be used in order to get accurate results. 
 In such a case, it is best to put 30 c.c. of distilled 
 water into a retort, add 10 c.c. of sulphuric acid and 
 10 c.c. of arsenious acid solution, then put in 10 c.c. 
 of the sample of water. Heat gently, and if only a 
 light blue colour is produced by a drop of the distillate, 
 when tested with the starch test, add other 40 c.c. of 
 the water to the retort, and proceed as above. This 
 method of estimation is exceedingly accurate with 
 small quantities of nitric acid, and with a water con- 
 taining much nitric acid, I have got perfectly accurate 
 
of Potable Waters 115 
 
 results when only using 10 c.c. of the sample of water 
 in the above manner. 
 
 Experiments illustrating the Accuracy of the Method. 
 
 Potassium nitrate 4*0 milligrms. = 0*554 milligrm. 
 nitrogen. 
 
 First experiment, used 4 milligrms. and obtained 
 0*576 milligrm. nitrogen. 
 
 Second experiment, used 4 milligrms. and obtained 
 0*543 milligrm. nitrogen. 
 
 Third experiment, used 4 milligrms. and obtained 
 0*552 milligrm. nitrogen. 
 
 Potassium nitrate 10 milligrms. 1*386 milligrms. 
 nitrogen. 
 
 First experiment, used 10 milligrms. and obtained 
 1*132 milligrms. nitrogen. 
 
 Second experiment, used 10 milligrms. and ob- 
 tained 1*124 milligrms. nitrogen. 
 
 The last two experiments show that the method 
 tends to be inaccurate (if carried out in the above 
 way) when the nitric acid is present in too large a 
 quantity. 
 
 The following series of waters had the nitrates and 
 
116 The Organic Analysis 
 
 nitrites estimated by the arsenious acid method, and 
 by the aluminium process. 
 
 Where aluminium was employed, 50 c.c. of the 
 sample of water and 40 c.c. of a 5 per cent, solution 
 of caustic soda (free of nitrates and nitrites) were 
 boiled down in an open flask to about 60 c.c. ; then 
 transferred to a tall glass cylinder (100 c.c. graduated 
 measure), and when nearly cold two strips of clean 
 sheet aluminium were added. The cylinder was 
 plugged with an india-rubber cork having a glass 
 tube passing through it containing glass-spun moist- 
 ened with pure sulphuric acid. At the end of about 
 twenty-two hours the contents of the cylinder and the 
 glass-spun were transferred to a distillation flask, and 
 the ammonia estimated in the usual way. 
 
 Nitrogen stated as grains 
 per gallon. 
 
 Arsenious acid. Aluminium. 
 
 Water Company, Maidstone 0'31 0'29 
 
 Well, Ashford Koad 1'98 2'15 
 
 Paradise Road 5' 76 5' 13 
 
 Bearsted 1'08 I'Ol 
 
 Eccles 4-27 4-11 
 
 Spout, Allington Lock 0-76 0'70 
 
 In every water, except one, the arsenious acid pro- 
 cess yielded more nitrogen than the aluminium process. 
 
of Potable Waters 117 
 
 This is what may be expected, as there is no certainty 
 about the aluminium having reduced all the nitric 
 acid, unless the water is treated with aluminium for 
 a second time; whereas with arsenious acid the absence 
 of the blue reaction with starch shows when the dis- 
 tillate no longer contains nitrous acid. The above 
 one exception is where the quantity of water used for 
 the arsenious acid method contained 6*9 milligrms. 
 of nitric acid. 
 
 APPENDIX E 
 
 Preparation of Solutions used in the Process for 
 estimating the Organic Carbon. 
 
 (1) Pure Sulphuric Acid. Add some powdered 
 permanganate of potash to some pure sulphuric acid 
 till the colour is moderately green, and then heat the 
 acid in a flask, covered with a beaker, till it becomes 
 colourless. 2 c.c. of the acid are used for each opera- 
 tion. 
 
 (2) Solution of Permanganate. Dissolve 10 grms. 
 of pure permanganate of potash in 1 litre of distilled 
 water, add 1 c.c. of pure sulphuric acid; boil the solu- 
 tion for one hour, and keep in a glass-stoppered 
 
118 The Organic Analysis of Potable Waters 
 
 bottle. 10 c.c. of the solution are used for each 
 operation. 
 
 (3) Solution of Ferrous Sulphate. Dissolve 50 
 grms. of the pure material in J litre of distilled 
 water, add 1 c.c. of pure sulphuric acid, and boil for 
 five minutes. Keep the solution in a glass-stoppered 
 bottle. 10 c.c. of the solution will decolorise more 
 than 10 c.c. of the permanganate solution. 
 
 (4) Centinormal Solution of Barium Hydrate. Dis- 
 solve 3' 15 grms. of crystallised hydrate in 1 litre of 
 hot water, boiled free of carbonic acid. This solu- 
 tion is centinormal for dibasic acids. Make the solu- 
 tion strongly pink with phenol-phthalein. 
 
 (5) Centinormal Solution of Oxalic Acid. Dissolve 
 1*26 grms. of the pure crystallised acid in 1 litre of 
 distilled water. This solution is centinormal for di- 
 acid bases. 10 c.c. of this solution will neutralise 
 generally more than 10 c.c. of the baryta solution, 
 because the baryta contains carbonate. 
 
 (6) Solution of Phenol-phthalein. Dissolve 0'5 
 grm. of the substance in 100 c.c. of distilled water, 
 aiding its solution by adding 5 c c. of rectified spirit. 
 This solution is pink in alkaline, and colourless in 
 acid or neutral solutions. 
 
 PBINTED BY ADLARD AND SON, BARTHOLOMEW CLOSE. 
 
UNIVERSITY Oi" 1 CALIFORNIA LIBRARY 
 
 THIS BOOK IS DUE ON THE LAST DATE 
 STAMPED BELOW 
 
 
 APR 30 192$ 
 
 30m-l,'15 
 
YB 16807 
 
 i-