LIBRARY 
 
 OF THE 
 
 'UNIVERSITY OF CALIFORNIA. 
 CTass 
 
. K 
 
LIME, MORTAR, HEMENT: 
 
 THEIR CHARACTERISTICS AND ANALYSES. 
 
 WITH AN" ACCOUNT OF 
 
 ARTIFICIAL STONE AND ASPHALT. 
 
 BY 
 
 W. J. DIBDIN, 
 
 Fellow and Member of the Council of the Inst. of Chemistry; Fellow 
 
 of the Chemical Society ; Member "of the Society of Public 
 
 Analysts; Member of the Society of ' Chemical 
 
 Industry ; Fellow of the Sanitary Institute. 
 
 Formerly Chemist and Superintending Gas Examiner to the 
 
 Metropolitan Board of Works and the London 
 
 County Council (1882 to 1897). 
 
 THE SANITARY PUBLISHING COMPANY, LTD., 
 5, FETTER LANE, E.G. 
 
 
 
J 
 
 GENERAL 
 
CONTENTS. 
 
 CHAPTER I. 
 
 CHARACTERISTICS OF VARIOUS LIMES. 
 
 PAGE 
 
 Pure Lime Chemical Character Solubility in Water Combina- 
 tion with Carbonic Acid Carbonate of Lime Sulphate of 
 Lime Hardness of Water Effect of Heat on the Carbonate 
 Effect of Water on Quick-lime Slaking Efflorescence 
 Composition of Typical Limestones Percentage Quantity of 
 Pure Lime and Magnesia in Properly Burnt Stone Fat or 
 Pure Limes Poor Limes Hydraulic Limes Hydraulic 
 Character of Limes Relative Uses of Different Limes 
 English and German Cements Specification (or Lime 5 
 
 CHAPTER II. 
 
 CLAY. 
 
 Action of Weathering on Rocks Silicate of Alumina Kaolin, 
 Marl, Loam Fire-clay Brick-clay Mudstone, Shale 
 Foul Clays Mild Clays Malm Pipe- clay Unctuous Clay 
 Red Clay Burnt Clay General Considerations Necessity 
 for Chemical Analysis 12 
 
 CHAPTER III. 
 
 SAND AND ITS SUBSTITUTES. 
 
 Silicon Natural Forms of Aluminium Silicates Magnesium 
 Silicates Mineralogical Characteristics Flint Mode of 
 Formation Water- worn and Sharp Sand Pit Sand River 
 Sand Sea Sand Differentiation of Quartz from Flint Arenes 
 "Chalk Sand " Grauwacke Rock - Standard Sand for 
 Cement Testing Leighton Bu/zard Sand Substitutes for 
 Sand Cinders Clinkers Puzzuolana Trass Use of by 
 Smeaton 17 
 
 104575 
 
IV 
 
 PAGE 
 
 CHAPTER IV. 
 
 MORTAR AND CEMENT. 
 
 Formation of Mortar and Cement Hydraulic Lime Mortar 
 Bye-laws of the London County Council Portland Cement 
 Manufacture of Burning to Clinker Fineness of Grinding 
 Mr. John Grant's Conclusions on Roman Cement Atkin- 
 son's Cement Medina Cement Selenitic Mortar P.'aster of 
 Paris Keene's Cement Parian Cement Stucco 24 
 
 CHAPTER V. 
 
 THE SETTING OF MORTARS AND CEMENTS. 
 Action of Silica Petzholt's Conclusions Degree of Carbonation 
 Crystallisation Precipitation Necessity for Clean, Un- 
 friable Sand and Grit. The Modern Mortar Mill Plaster of 
 Paris Effect of Heat on a Solution of Supersaturated Solu- 
 tions Influence of the Character of Sand Effect of Heat on 
 Suggestions for Investigation Mr. John Grant on Cohesive 
 and Adhesive Strengths Necessity for Using Wet Bricks ... 30 
 
 CHAPTER VI. 
 
 ROUGH TECHNICAL TESTS OF MORTAR WITHOUT CHEMICAL 
 ANALYSIS. 
 
 Practical Experience of Mortar Necessity for Analysis in Certain 
 Cases Rough Test for Strength Vegetable Debris 
 Kind of Grit and Relative Proportions of Lime in Mortar 
 Earthy Matter Setting Power Use of such Tests in Disputes 37 
 
 CHAPTER VII. 
 
 THE CHEMICAL ANALYSIS OF LIME, MORTAR, CEMENT, &C. 
 
 Lime and Limestone Preparation of Sample Constituents to be 
 Looked for Procedure Moisture Silica Iron Alumina 
 Volumetric and Gravimetric Processes Lime Magnesia 
 Alkalis Carbonic Acid Mortar Preparation of Sample 
 Sand and Grit Analytical Procedure Clays Alluvial Muds 
 Portland Cement Keates' Specific Gravity Bottle Fre- 
 senius* Specific Gravity Bottle and Method of Examination of 
 Cements Conclusions Based thereon Tabulation of Results 
 Adulterated Cement Interpretation of Fresenius' Results... 43 
 
PAGE 
 
 CHAPTER VIII. 
 
 TYPICAL ANALYSES OF MORTAR AND CEMENT. 
 Legal Specification of Mortar Usual Character of Mortar 
 Earthy Matter Soluble Silica Carbonic Acid Typical 
 Analyses of Lime and Brick Mortar and Ingredients Con- 
 version of Weight of Commercial Lime to a Volume Corre- 
 sponding to that of Broken Brick Weights of One Cubic Foot 
 of Different Materials Fresh Limes and Cements Relation 
 of Weight to Volume (Cement Mortar) Soluble Silica 
 Absorption of CO 2 by Mortar Tables of Typical Analyses of 
 Good and Bad Lime and Cement Mortars Discussion at the 
 Society of Public Analysis ... ... 66 
 
 CHAPTER IX. 
 
 MECHANICAL TESTS OF CEMENT. 
 
 Specification for Portland Cement, &c.~- American and Continental 
 Standards of Fineness and Tensile Strength Mr. Grant's 
 Conclusions Mechanical Tests Sampling Gauging and 
 Manipulation Fineness Expansion or Contraction, &c. 
 Mr. Faija's Results Faija's Cement Testing Machine Salter 
 and Co.'s Compound Lever Cement Testing Machine Per- 
 colation Tests Messrs. Hilton and Co.'s Apparatus 85 
 
 CHAPTER X. 
 
 THE ADULTERATION OF PORTLAND CEMENT 
 Questionable Practices Conclusions of the London Chamber of 
 Commerce, and of the Union of German Cement Manufac- 
 turers Admixture of Gypsum, Effect of Fraudulent Admix- 
 tures Experiments of Messrs. Stanger and Blount Reso- 
 lution of London Chamber of Commerce thereon Stanger 
 and Blount's Report Tensile Tests of Cements containing 
 Ragstone Analyses of Samples Compressive Tests Neat 
 Tests Sand Tests Results of Aeration of Cement Micro- 
 scopical Investigation Tests for Constancy of Volume Con- 
 clusions as to Addition of Ragstone Additions to Cement 
 other than Ragstone Gypsum Blast Furnace Slag Erd- 
 menger and Lundteigen on Various Admixtures with both 
 Fresh and Salt Water 104 
 
VI 
 
 PAGE 
 
 CHAPTER XI. 
 
 THE STRENGTH OF BRICKWORK. 
 
 Science Standing Committee to the Institute of British Architects 
 Experiments on Brick Piers Varieties of Bricks Used 
 Crushings at Three and Nine Months Experimental Plant 
 Composition of Mortar employed Mr. Matt. Garbutt's Notes 
 on Results Tables of Collapsing Pressures Prof. Unwin's 
 Tests of Bricks, Lime, Cement, and Sand Used Analyses by 
 Author Second Series of Tests Committee's Observations 
 Impressions of Committee thereon Brick Walls, Table of 
 Collapsing Pressures Committee's Observations on the "Safe 
 Load" 134 
 
 CHAPTER XII. 
 
 CONCRETE. 
 
 Definition Betor. Historical Notfs Chinese Methods Roman 
 Methods Concrete in Building Construction Mixing De- 
 positing Tests of Compressed and Uncompressed Concrete 
 Blocks, Mr. Grant's Results Ramming Effects of Tempera- 
 ture and Atmospheric Influences Hastening the Setting of 
 Concrete Effect of Ferrous Sulphate Puscher's Results 
 Laudrin's " Pouzzo-Portland " Specification for Concrete 
 Boulnois on Street Foundations 145 
 
 CHAPTER XIII. 
 
 ARTIFICIAL STONE. 
 
 History of Ranger's Patents Coignet's " Beton Agglomere "- 
 Ransome's " Siliceous Stone " Buckwell's "Granitic Breccia " 
 The " Victoria Stone " Composition of Granite Used 
 Com position of the "Stone" Ward s Stone Stuart's "Grano- 
 lithic Stone" "Non-slip" Stone Moreau "Marble" 
 Terra-cotta Composition of Clay Used Aichitectural Appli- 
 cations General Observations 158 
 
 CHAPTER XIV. 
 
 ASPHALT. 
 
 Investigations of Mr. Clifford Richardson Trinidad Lake Asphalt 
 Experimental Borings Description of Deposit Variation 
 
VI 1 
 
 PAGE 
 
 in Levels Influx of Pitch at the Present Time Movement of 
 the Lake Surface Proximate Composition of Trinidad 
 Asphalt Methods of Analysis Average Composition Bitu- 
 men Soluble in Naphtha Deposits of Land Asphalt Proxi- 
 mate Composition and Properties Crude Lake Pitch Soft 
 Pitch and Pitch from Blow Holes Water in the Pitch and in 
 the Lake Mineral Matter in the Pitch The Bitumen of 
 Trinidad Asphalt The Bermudez Asphalt Deposit Harden- 
 ing of the Main Mass of Pitch Methods of Analysis Sulphur 
 in Asphalt and its Influence Ultimate Composition Uvalde 
 County Bitumen " Pittsburg Flux " Action of Acids on 
 Asphalts Naphtha Soluble Bitumen in various Asphalts 
 Effect of Heat on Mineral Oils Conclusions as to the Nature 
 and Origin of Asphalt Chinese and Egyptian Bitumen Rate 
 of Influx of Asphalt into the Pitch Lake 174 
 
INTRODUCTION. 
 
 THE following short nrrni\nf nf fhf* rhararfpr nn/I 
 
 ADDENDA AND CORRIGENDA. 
 
 On page 20, 6th line from bottom, read "Arenes" 
 instead of "arcues." 
 
 On page 22, 2nd line from bottom, read "Andernach" 
 instead of "Audernach." 
 
 On pages 22 and 23, read "pozzuolana" instead of 
 "puzzuolana." 
 
 Page 49, 2nd line from the bottom, should read: "B = 
 the total weight of the. chlorine found in the 
 chlorides. The " 
 
 ot the setting of mortar and cement, a question which is not 
 yet finally solved The various theories are discussed in 
 the chapter relating to that subject, and some general 
 suggestions made for further consideration. 
 
INTRODUCTION. 
 
 THE following short account of the character and 
 chemical composition of the various materials used in 
 making mortars and cements, and the methods and results 
 of their chemical analysis, has been compiled to meet the 
 requirements of those who in the course of their technical- 
 experience have hitherto been at a loss where to look for 
 .guidance. 
 
 The author's experience has shown the necessity for 
 recognised methods of examination and authentic results, 
 which may be looked upon as a basis in the disputes which 
 constantly occur in regard to the purity and strength of 
 these materials. 
 
 In the following pages an account of .the various limes 
 will be found, the importance of which alone is very far 
 reaching. By the detection of improper ingredients, and 
 the establishment of a proper specification, the author was 
 enabled on one occasion to remedy very grave defects, and 
 to effect a saving to one firm alone of no less a sum 
 than ^2000 per annum ! The various kinds of clay and 
 sand, and their substitutes, are also fully dealt with. 
 
 Much dispute has arisen at various times as to the cause 
 of the setting of mortar and cement, a question which is not 
 yet finally solved. The various theories are discussed in 
 the chapter relating to that subject, and some general 
 suggestions made for further consideration. 
 
The analytical qjestion has been, the author trusts, fully- 
 met, and the chem cal and physical factors of the problem 
 clearly set out in detail. The results have been obtained 
 only after much careful work, in connection with which 
 sincere thanks are due to the painstaking work of Mr. R. 
 Grimwood, F.I.C., official Chief Assistant in the Chemical 
 Department of the London County Council. 
 
 The rough technical tests of mortar have been found by 
 the author to be of no little value when time presses, and a 
 general idea of the quality of the sample is required at 
 short notice. 
 
 The mechanical tests for cements in particular are fully 
 described, as no work of this kind would be complete 
 without such a description, the author, however, does not 
 claim any special features for these, which are so well 
 known, but doubtless the opinions of the various experts- 
 quoted will be found of great assistance. 
 
 It is with peculiar pleasure that the results of the- 
 exhaustive researches of the author's former colleague, Mr. 
 John Grant, C.E., on the testing and use of Portland 
 cement, are included. It is unnecessary to do more than 
 to refer to these classical researches to ensure their due 
 share of attention. 
 
 The adulteration of Portland cement has of late years- 
 been brought prominently to the front owing to the position 
 taken up by the London Chamber of Commerce, which 
 has spoken plainly on the question. In this connection the 
 researches of Messrs. Stanger and Blount will be studied 
 with no little interest. 
 
 The elaborate experiments on the strength of brickwork 
 
undertaken by a Committee of the Institute of British 
 Architects need little recommendation. This invaluable 
 work, by the kind permission of the Institute, has been 
 carefully summarised in the following pages. 
 
 The use of concrete has now taken such a place in the 
 engineering work of the day, that no apology is needed for 
 the inclusion of a chapter on that interesting subject. 
 Following this will be found an account of the various 
 artificial stones now so largely employed. 
 
 The author is greatly indebted to Mr. Clifford Richard- 
 son, of Long Island City, New York, for his kind 
 permission to insert copious extracts from the account of 
 his valuable researches on the nature and origin of asphalt. 
 Probably no other work contains such an exhaustive and 
 reliable account of this valuable material. 
 
 The author has to express his sincere thanks also to Mr. 
 Messrs. Stanger and Blount, Mr. C. Chambers Smith, C.E., 
 Mr. R. Grimwood, F.I.C., Mr. R. G. Grimwood, A.I.C., 
 Mr. Searles-VVood, F.R. Inst. B.A., and many other friends, 
 for their assistance and kind permission to embody much 
 material, without which the work would have been of far 
 less value than it is hoped it will -be. 
 
 It will be observed that the work has been condensed as 
 far as possible with a view to greater convenience for ready 
 reference. 
 
 13 2 
 
CHAPTER I. 
 
 CHARACTERISTICS OF VARIOUS LIMES. 
 
 LIME in its pure state is the oxide of calcium. 
 It is white, caustic, soluble in water to the extent of 
 1 1 grains in a pint, or 88 grains per gallon, and it combines 
 
 with carbonic acid to 
 form carbonate of lime,, 
 which is familiar to us 
 in an impure state in the 
 form of chalk. The 
 metal calcium is never 
 met with in an uncom- 
 bined state in nature, 
 being always in com- 
 bination, chiefly exist- 
 ing as carbonate. The 
 "atomic" weight of the 
 metal is forty, compared 
 with hydrogen taken as 
 unity. In its pure state 
 it is yellow, harder than 
 
 A CHALK QUARRY. ^ duct j le an( J ma j. 
 
 leable. It combines with oxygen slowly in dry air, more 
 rapidly in the presence of moisture, and must be preserved 
 out of contact with air. When heated in air to a bright 
 red heat it fuses and ignites, burning with a bright white 
 light, and forming the oxide of calcium. Calcium 
 oxide is the pure substance of our inquiry, but, as 
 may be supposed, it is never obtained absolutely pure in 
 commerce in fact, for the purpose of making mortar and 
 
cement it would be of but little use unless it had mixed 
 with it other materials, such as silica, iron, and alumina. 
 
 In the form of carbonate, calcium forms a principal 
 portion of some of the most common minerals and rocks, 
 viz., arragonite, calc-spar, chalk, limestone, marble. It is 
 a constituent of dolomite, which is a compound of carbonate 
 of lime and carbonate of magnes : a, having the composi- 
 tion CaCo;. MgCo 3 . Calcium also occurs as sulphate in 
 anhydrite (CaSo 4 ), in gypsum, alabaster, and selenite 
 (CaSo 4 , 2 H,O) ; as fluoride, in fluorspar (CaF,) ; a 
 phosphate in apatite [CaF 2 , 3 Ca 3 (PO 4 )J ; and as silicate 
 is widely diffused in different minerals. Many of the 
 salts of lime are normal constituents of soils, plants, 
 and of the animal body ; egg shells, coral, and the shells 
 of the mollusca consist chiefly of calcium carbonate, whilst 
 the mineral matter of bones consists chiefly of calcium 
 phosphate. The carbonate (held in solution by free 
 carbonic anhydride, better known as carbonic acid) and the 
 sulphate are also normal constituents of natural waters, 
 and cause, with the magnesia and other salts, the property 
 known as hardness. 
 
 If a fragment of white marble, which is practically pure 
 carbonate of lime, be heated in such a manner that the 
 carbonic acid set free can escape, a white substance will 
 be obtained, which is the oxide of calcium, or quicklime, 
 having the composition, chemically expressed, of CaO, 
 thereby indicating the combination of one atom of calcium 
 with one atom of oxygen, and this compound has a 
 " molecular " weight of 56, being 40 for the calcium and 
 1 6 for the oxygen. Its specific gravity is 3^08, water being 
 taken as unity. 
 
 This white, amorphous solid attracts water and carbonic 
 acid from the air, and under favourable conditions combines 
 
with water with great energy and much evolution of heat, 
 forming calcium hydroxide, according to the equation, CaO + 
 HX) = Ca (OH);,. When water is poured upon quicklime, the 
 latter swells up, cracks, and emits steam, and finally falls into 
 .an impalpable powder. The purer arid fresher the lime the 
 more rapidly does this phenomenon occur. In this condition 
 the lime is known as "slaked/"' If an excess of water has not " 
 been employed, the powder is to all appearance dry, ancti&iar : 
 finer^ than the unslaked lime can be obtained by any process 
 of grinding. When in the course of time the lime again 
 combines with carbonic acid by absorbing it from the atmo- 
 sphere, it necessarily parts with the water so taken up, the 
 carbonic acid replacing it; thus, Ca (OH) 2 + CO. = CaCO 3 + 
 PLO. This action is the cause of the efflorescence on brick- 
 work known as " salting out," which will be referred to 
 later on. The water thus displaced dissolves other salts, 
 .and the solution formed finds its way to the surface of the 
 work, where the water evaporates, leaving the crystallised salt 
 as a crust on the surface of the wall. From .the above observa- 
 tion it will be seen that lime, considered as a chemical sub- 
 stance, is a very definite body, having characteristics peculiar 
 to itself, and that these characteristics are the same wherever 
 the lime may have been obtained. It is when we come to 
 consider the actual properties of the various limes obtained 
 in commerce that certain very specific characters are brought 
 into play, and it is these which we have to consider in rela- 
 tion to the adaptability of the commercial lime for certain 
 uses. Thus the lime most suitable for making mortar is by 
 no means the best for making lime-water in large quantities, 
 or for the purpose of the gas manufacturer in purifying his 
 gas. These differences, it will be seen, are due to certain 
 substances or impurities, strictly speaking which affect 
 the quality of the lime. 
 
The following table showing the approximate composition* 
 of some typical limestones, and the percentage quantities of 
 pure lime and magnesia in the properly burnt stone, wilj 
 
 Table of the approximate composition of various Limestones, and the 
 percentage of pure Lime and Magnesia in the Burnt Stone. 
 
 Limestone. 
 
 Unburnt Stone. 
 
 In Burnt 
 Stone. 
 
 Carbonate 
 of Lime. 
 
 Carbonate 
 of Magnesia. 
 
 r u 
 
 II 
 
 0<j CO 
 
 j?| 
 
 o 
 
 Lime CaO. 
 
 Magnesia 
 MgO. 
 
 PUKE, OR " FAT," LIMES 
 
 
 
 
 
 
 Marble 
 
 100 
 
 
 
 
 
 IOO 
 
 
 
 White Chalk 98^ 
 
 o 
 
 I 
 
 98 
 
 o-i 
 
 Oolite ! 95 
 
 2 
 
 3 
 
 95i 
 
 *i 
 
 POOR LIME 
 
 
 
 
 
 
 Silicious Oolite 
 
 70 
 
 3i 
 
 26J, 
 
 73j 
 
 3i 
 
 FEEBLY HYDRAULIC 
 
 
 
 
 
 
 Grey Chalk 
 
 92 
 
 
 
 8 
 
 85J 
 
 
 
 HYDRAULIC 
 
 
 
 
 
 
 Dolomite 
 
 5 1 
 
 40 
 
 9 
 
 53i 
 
 36 
 
 Carboniferous 
 
 86 
 
 
 
 H 
 
 81 
 
 
 
 Grey Chalk 
 
 83 
 
 
 
 T 7 
 
 73 
 
 
 
 STRONGLY HYDRAULIC 
 
 
 
 
 
 
 Blue Lias 
 
 79 
 
 - 
 
 21 
 
 72 
 
 
 
 Carboniferous 
 
 ?ri 
 
 'i 
 
 27 
 
 60 
 
 I 
 
 Scotch 
 
 68 
 
 i 
 
 31 
 
 56* 
 
 o-i 
 
 serve to indicate the differences in character of the raw- 
 material before it is converted into lime by being burnt in 
 the kiln. First of all, it is to be noted that the various. 
 
limestones may be classified under four heads, viz. : (i). 
 Fat or pure limes ; (2) poor limes ; (3) hydraulic limes; (4} 
 cements, natural and artificial. As it will be seen, these 
 classes merge gradually into one another, and it is difficult 
 to draw a sharp line of demarcation in many cases : there- 
 fore we have "good" and "bad" limes, according to the 
 purpose for which they are intended. These analyses 
 may be accepted as forming an accurate sequence represen- 
 tative of the limes in question. 
 
 From this table it will be seen that the hydraulic character 
 of some limes is due to the presence of certain impurities, viz.,. 
 silica, iron, and alumina, which are included under the head 
 of "clay "; and that as these increase in quantity up to the 
 extent of about 30 per cent, the " hydraulicity," as it is 
 called, of the lime produced by the burning of the stone is 
 increased. 
 
 It will now be seen how important it is that the use to- 
 which the lime is to be put should be first considered before 
 a purchase is made. Thus, if the lime is required for use 
 in a gasworks, &c., where the chemical activity of the lime 
 is of the first importance, it should be as pure as possible, 
 and therefore " fat " limes should be obtained. If, on the 
 other hand, the lime is required for mortar making, good 
 hydraulic lime should be purchased. In the case of the 
 manufacture of cement, the quality of the limestone employed 
 must be considered in regard to the quantity of clay to be 
 mixed with it to form the " slurry," from which the cement 
 is manufactured. This point serves to indicate the import- 
 ance of chemistry in regard to our manufactures. For 
 years the English Medway cements were considered far 
 superior to that of German manufacture, because the in- 
 gredients naturally to hand constituted the best materials 
 for the purpose ; but as soon as the German manufacturers 
 
IO 
 
 -studied the question from the chemical standpoint they 
 ^ere enabled to so choose their ingredients as to turn out 
 cement equal to the best of the English makes. 
 
 It will be understood that when the limestones are burnt 
 the relative proportions of lime, &c., vary by reason of the 
 loss of the carbonic acid which is driven off in the process 
 Thus, while the marble would show still 100 per cent, of 
 .pure lime, there being no impurity present, the white chalk 
 would yield a lime having the composition of: Lime, 55-2 
 clay, i'2 = 56*4; or, expressed in percentages, lime, 97*9; 
 clay, 2 'i. In like manner the carboniferous limestone, 
 having 86*2 per cent, of carbonate of lime, would, by the 
 loss of carbonic acid, yield only 48*3 of lime to 11*2 of clay 
 the percentages being: Lime, 81*2; clay, 18-8. It will 
 be remembered, from what has been stated in the first part 
 of the chapter, that each 100 parts of carbonate of lime 
 yields, if thoroughly burnt, only 56 of lime, by reason of the 
 loss of 44 parts of carbonic acid. 
 
 The following specification for the supply of lime has 
 been found by the author to fully answer its purpose, and 
 may usefully be inserted here : 
 
 GENERAL CONDITIONS. 
 
 T. The lime is to be delivered by the contractor at the 
 works, alongside the wharves or jetties of the purchaser, as 
 may be directed from time to time. 
 
 2. Each parcel is to be sampled on delivery, and in the 
 first instance examined by the purchaser. In the event of 
 his analyses differing from those of the sellers, samples are 
 to be submitted to a referee, whose report shall be binding 
 on both parties. Should the contractor not be represented 
 at the time of the sampling (of which due notice will be 
 
II 
 
 .given) the decision of the purchaser as to the quality cf the 
 lime must be accepted as final. 
 
 3. The lime is to be clean, well-burnt, hand-picked 
 lime, free from dust, coke, and clinker The price quoted 
 by the tenderer is to be for lime containing 95 per cent, of 
 actual caustic lime (CaO). The purchaser may reject any 
 parcel of lime containing less than 85 per cent, of actual 
 caustic lime, but may, if he think fit, accept the \\hole or 
 any part of such parcel. 
 
 The price to be paid for any lime delivered and accepted, 
 which contains more or less than 95 per cent, of actual 
 caustic lime, is to be greater or less than the price quoted 
 in strict proportion to the excess or deficiency of actual 
 caustic lime, except that where the deficiency exceeds 10 
 per cent., the deficiency beyond 10 per cent, is, in fixing 
 the price as aforesaid, to be considered as double what it 
 .actually is. 
 
 4. Samples of the lime proposed to be delivered, with 
 the name of the source or district from which the same is 
 obtained, must be submitted with each tender, which will 
 otherwise not be considered. 
 
12 
 
 CHAPTER II. 
 
 CLAY. 
 
 WEIGHING OUT THE CLAY. 
 
 BY the process known as " weathering," already formed 
 rocks are gradually disintegrated, and the fine mud thus 
 formed constitutes the very variable materials classed for 
 convenience under the generic term Clay. The principal 
 ingredient is a silicate of alumina, which when perfectly 
 pure is white. In this form it is known as Kaolin, or white 
 china-clay, pipe-clay, fuller's earth, &c. These correspond 
 more or less to the composition A1,O 3 , 2 Si(X, 2 H 3 O.. 
 White marl, loam, c., are classes containing various- 
 
'3 
 
 admixtures, such as calcium carbonate, oxides of ircn, 
 magnesia, c. . 
 
 Fire- clay ', largely found in the proximity of coal seams, con- 
 tains iron, and is nearly free from lime and the alkalies, soda, 
 and potash. These are largely used for the manufacture of 
 pottery, as at Stourbridge. The most refractory, used for 
 the manufacture of articles which have to stand an intense 
 heat, is composed of: Silica, 74 per cent; alumina, 16 per 
 cent; ferrous oxide, 3 per cent.; alkalies, i per cent.; water, 
 6 per cent.; with traces of lime, magnesia, sulphuric acid, 
 and chlorine. 
 
 Brick-clay is a term applied industrially to any variety of 
 clay, loam, c., used for making bricks and low-class 
 pottery. It is very impure, containing large quantities of 
 iron, c. The following may be taken as its approximate 
 composition, but various samples differ very largely : 
 Silica, 50 per cent.; alumina, 35 per cent.; ferric oxide, 
 8 per cent.; lime, i to 2 per cent; magnesia, 5 per cent., 
 &c. The colour of the bricks is due to the quantity of 
 iron present, and the extent to which the bricks are burnt. 
 
 Mudsione is a fine sandy argillaceous rock, free from 
 stratification, and harder than the usual clays. 
 
 Shale is a clay which has beeri deposited in thinly 
 stratified layers. These are capable of being split into hard 
 sheets or leaves. There is a great variety of these deposits 
 having very different characters, according to the nature of 
 the clay originally deposited, gradually passing into the 
 slates. 
 
 In the technical terms of the brickmaker, pure clays, 
 composed of silicate of alumina, with but small quantities ot 
 lime, magnesia, c., are known as "foul" clays. Loams, 
 or "mild" clays containing sand, are known as "sandy 
 clays." 
 
J4 
 
 Marl is a clay containing a large proportion of carbonate 
 of lime. 
 
 Malm is an artificial imitation of natural marl, made by 
 mixing clay and chalk in a wash-mill. It is generally 
 known under the term " washed clay." 
 
 Pipe-day is a greyish-white clay, greasy to the touch, and 
 adheres strongly to the tongue. On being burnt it turns 
 white. It is found in Devonshire, Dorsetshire, and on the 
 Continent. It is used for the manufacture of clay pipes 
 hence its name. 
 
 [A species of unctuous clay is eaten by the Ottomaques, a 
 tribe near the Orinoco. This habit, doubtless originating 
 in famine, occurs also in Brazil and other places. It is 
 stated that the clay used has a milky and not disagreeable 
 taste.] 
 
 Red Clay is distributed to an enormous extent over the 
 bed of the ocean, and may be said to exist everywhere in 
 the form of ooze, clay, or mud. It is generally of a 
 brown colour, in consequence of the presence of the oxides of 
 iron, with some manganese. It has not the same properties; 
 as fire-clay, as it fuses into a magnetic bead, due to the 
 large quantity of iron and free silica, derived from the 
 sedimentation of the skeletons of silicious organisms, such, 
 as the diatomaceae, and volcanic minerals. 
 
 Burnt Clay is very largely used for many purposes. As; 
 a manure it has, in certain cases, considerable value, whilst 
 it is in great favour with the builder and engineer for many 
 purposes. Within the last few years the use of burnt clay has- 
 been extended to the sanitary world, being employed for 
 filling the " Bacteria Beds," which are now so rapidly dis- 
 placing the chemical and land treatment for the purification 
 of sewage. The method of burning the clay is to first 
 kindle a fire with suitable materials round a stout piece of 
 
oak or other hard wood. The clay is gradually piled up- 
 round this, and sprinkled with fine coal slack or breeze. 
 Care must be taken that the fire does not break through 
 the surface, as in that case over-heating may result, unless 
 the burnt clay is required for certain purposes, for which it 
 cannot be too well burnt. The alternate layers of clay and 
 breeze are gradually added from time to time until the heap 
 may equal hundreds of tons. Constant attention is required 
 to see that the firing is equally distributed, and kept well 
 within the heap. 
 
 Probably few materials have contributed more to the 
 commercial success of the English industries than the 
 wealth of good clays of all kinds which abound in this 
 country. Other materials may show a larger net return, 
 but in nearly all these cases clay in one form or another has 
 been employed, even if only for the buildings in which the 
 various operations are conducted. To clay we owe our 
 buildings, our furnaces, our railways and their embank- 
 ments and bridges, our potteries. It is largely used by our 
 paper-makers, and in an endless variety of the arts and 
 industries, contributing in a multitudinous number of ways, 
 to our health, our comfort, and our necessities. 
 
 From the foregoing short account of the various descrip- 
 tions of clay which exist, it will easily be seen how important 
 it is to ascertain exactly the character of any sample which 
 it is proposed to use for the manufacture of any special 
 article. In the case of cement it is clearly of the greatest 
 importance to ascertain that the right proportions of silica 
 and alumina are present. If, for instance, the cement 
 maker has only at his disposal a marly clay, it is useless for 
 him to proceed as if he were using a "foul " clay, as the pro- 
 portions of lime to silica and alumina would be very- 
 different. 
 
i6 
 
 In like manner a mild or loamy clay must have the pro- 
 portions of limestone increased, or it may be that the 
 quantity of sand present would be excessive, and thus render 
 it unsuitable. 
 
 It is in this connection that the value of a properly con- 
 ducted chemical analysis is felt. In these days of severe 
 competition the manufacturer cannot profitably neglect the 
 ;aid of science, which has aided so much those industries 
 which have been controlled by those who are wise enough 
 to avail themselves of the help which precise knowledge can 
 .give. 
 
CHAPTER III. 
 
 SAND AND ITS SUBSTITUTES. 
 
 NEXT to oxygen, silicon is the most abundant and widely- 
 diffused element known. It never occurs in the free state, . 
 but always in combination with oxygen as silica (SiO 2 ), 
 either alone or in combination with other substances, thus 
 forming a large series of minerals, of which the following 
 are some of the natural forms in which it is found : 
 
 Quartz Opal 
 
 Amethyst Jasper . 
 
 Agate Flint 
 
 Chalcedony Sandstone 
 
 Onyx Infusorial earth 
 
 Cornelian 
 
 Quartz and amethyst occur in well-shaped crystals,, 
 belonging to the hexagonal system (six-sided prisms). 
 Agate and chalcedony are a mixture of crystallised and 
 amorphous silica. 
 
 The following compounds of silica may be broadly classed 
 as aluminium and magnesium silicates, but most of them 
 contain other elements, such as sodium, potassium, calcium,, 
 iron, manganese, c. Their chemical constitution is very- 
 complicated, and it is very doubtful as to whether it is 
 thoroughly known : 
 
 Aluminium Silicates. Magnesium Silicates. 
 
 Beryl Asbestos 
 
 Clay Augite 
 
 Felspar Hornblende 
 
 Garnet Meerschaum 
 
 c 
 
i8 
 
 Aluminium Silicates. Magnesium Silicates. 
 
 Jade Olivine 
 
 Lapis Lazuli Steatite (French Chalk, 
 
 Mica Soapstone, &c.) 
 
 Pumice Stone Talc 
 
 Slate 
 
 Topaz 
 
 The mineralogical characteristics of these substances are 
 of considerable interest. The quartz occurs in various forms, 
 such as rock crystal, which is transparent and colourless ; 
 rose quartz, red inclining to violet blue ; milk quartz, 
 milk white, and slightly opalescent cat's eye, enclosing 
 asbestos, greenish white or grey, olive green, red, brown, or 
 yellow ; avanturine, enclosing mica, yellow, red, green, or 
 brown ; siderite, indigo or Berlin blue. 
 
 Common quartz, either crystallised or massive, of various 
 colours, is a constituent of many rocks ; ferruginous quartz, 
 or iron flint, is often found with iron ores. 
 
 Flint is of various colours, from greyish-white to red or 
 brown. It is semi-transparent, and has a flat conchoidal 
 fracture. It is found chiefly in the chalk formations and 
 river drifts. The mode in which flint is formed is assumed 
 to be by the gradual accumulation of soluble silica upon the 
 silica forms of minute marine organisms, as traces of these 
 are almost always found in the flint nodules. Dr. Bower- 
 bank was of opinion, from the prevalence of silicious 
 spicules of sponges in the nodules, that all flints had for their 
 primary nuclei the silicious framework of the sponges which 
 flourished in the depth of the sea during the cretaceous 
 period. Flints are found often in the chalk hollows in ex- 
 tensive layers of beds, such as that in the Chipstead Valley, 
 in Surrey, where, it is said, traces of ancient workings have 
 
been found, and where at this time large numbers of flint 
 stones are still unearthed. 
 
 Sand is composed of either the simple fragments of quartz 
 and flint which have undergone but little trituration by the 
 action of flowing water in rolling the particles together, or 
 similar fragments after they have been rubbed together for 
 long periods, whereby the sharp edges have been ground 
 down, and the particles become more or less rounded. In 
 the former condition, the sand is known as " sharp " sand, 
 whilst in the latter it is " water- worn " sand. 
 
 fit Sand is of the former character, and is the best 
 suited for making mortar, but it often is mixed with clay 
 and other matters. 
 
 River Sand belongs to the water-worn variety, and is 
 generally very clean, fine, and white. 
 
 Sea Sand is, by reason of the greater attrition which it 
 has undergone, very round and smooth. Its use is objected 
 to by some as giving rise to the phenomena known as " salt- 
 ing out " after the work is finished. This is doubtless due 
 to the sand not having been washed -in fresh water to 
 remove the salts with which the sea water is so highly charged, 
 and which, drying with the sand, aftarwards crystallise out 
 on the face of the work. 
 
 In connection with the investigation as to the deposits in 
 the riv'er Thames by the Royal Commission presided over 
 by the lite Lord Bramwell, in 1883, Dr. Sorby pointed out 
 the following beautiful method of differentiating the 
 particles of quartz from those of flint in the sand of the 
 river bed. 
 
 Quartz sand is made up either of entire crystals or of 
 fragments of crystals, so that each particular grain, has one 
 certain optical structure. A grain of quartz sand examined 
 .by polarised light under the microscope will appear to be 
 
 c 2 
 
20 
 
 black when the two prisms of the polariscope and the grain 
 of quartz are in a certain position to one another. It is not 
 shown on the field of the microscope, because the axes of 
 depolarisation are parallel. As the quartz is rotated on its 
 axes, without the prisms being altered in their position, it 
 becomes brilliantly illuminated and coloured with all the 
 prismatic colours, and as the rotation is continued these 
 subside, and the quartz is once more in darkness ; and this 
 alternation of brilliant illumination and darkness continues 
 as the rotation brings the axes of polarisation parallel or not 
 with that of the prisms of the polariscope. 
 
 Flint is made up of a great many very small crystals, and 
 when it is seen in polarised light looks mottled. The differ- 
 ence between flint and quartz corresponds to the difference 
 between a mass of granite and a piece of coloured glass. On 
 the whole, the flint would look mottled, with black and white 
 all over. The bright colours are not seen, because the size 
 of the individual crystals is not sufficiently great to produce 
 interference of a sufficiently high order to give any colour. In 
 a hundred grains or so, one might be met with which 
 would cause doubt, but the differences between the quartz 
 sand and flint sand is as well marked as the difference 
 between sheep and goats. 
 
 In France a kind of sand exists in large quantities, 
 to the extent even of hills, which consist of finely- 
 divided chalk in grains of various sizes mixed with 
 various coloured clays in different proportions. That is 
 called " arcues," and is used with the rich loam from 
 the rivers. Disintegrated Grauwacke rocks afford an 
 argillaceous sand, consisting of quartz, schist, felspar, mica, 
 &c., held together by clay. This is used both in its natural 
 condition, and also after calcination, when it resembles 
 Duzzuolana, referred to later on. 
 
21 
 
 Standard Sand for Cement Testing. In consequence 
 of the great variety in the results obtained in the tests of 
 Portland cement with different descriptions of sand, it was 
 found desirable to agree to some uniform method of pro- 
 cedure. This was pointed out by the late Mr. John Grant 
 very clearly. In the course of his researches he found that 
 " The washing of the sand seems to add about 35 per cent, 
 to the strength at the end of a week ; but at the end of a 
 month this experiment is not so satisfactory. With washed 
 pit sand, in the proportion of one of cement to two of sand, 
 the breaking weight, at twenty-eight days, was 330 lb., or 
 46 *2 per cent, above the first specimen of Thames sand, or 
 31*21 per cent, above the same sand when not washed."* 
 
 The result of these and numerous other similar experi- 
 ments was that a general agreement was arrived at to 
 employ Leighton Buzzard sand, which is first carefully 
 washed to remove all earthy matter, <Scc., and then sifted 
 through a mesh of 400 to the square inch, and again 
 through a mesh of 900 to the square inch The portion of 
 the sand taken as a standard for cement testing is that 
 which is caught on the 900 mesh sieve. 
 
 Substitutes for Sand. Whilst crushed flints or quartz, by 
 reason of the sharp edges giving a better hold to the brick- 
 work, is by far the best material to use with lime for making 
 mortar, well-washed pit sand or river sand are the next best 
 materials for the purpose. Many substitutes for these have 
 been proposed. Chief amongst these may be placed well- 
 iburnt clay, either in the form of ballast broken down to a 
 sufficient degree of fineness, or as broken bricks. The 
 Romans are said to have used pounded tiles very largely, 
 and the French are strongly in favour of it. When the 
 
 * "The Strength of Cement." J. Grant. Page 8. 
 
22 
 
 mortar was made with fresh water the results were ver 
 satisfactory, but when exposed to the action of sea water the- 
 work lasted only a few years. In England crushed burnt 
 clay and bricks are very largely used, but unfortunately, the 
 system of grinding all the materials together in a mill lends 
 itself to the introduction of large quantities of earthy 
 matter, which has a most disastrous effect on the subsequent 
 strength of the mortar. 
 
 Well-burnt cinders or clinkers make excellent mortar,, 
 whilst furnace slag and the scoria) from ironworks answer 
 very well. The use of coal cinders from the domestic dust- 
 bin is most strongly to be deprecated, unless these are 
 passed through a "destructor" to destroy all the foul 
 organic matter with which they are unfortunately so freely 
 mixed. The author has examined many samples of mortar 
 so prepared, which have been after a short time putrid and 
 covered with fungoid growths. The use of such materials, 
 unfortunately only too common with unscrupulous builders,, 
 should be made a criminal offence, punishable with 
 imprisonment. To sacrifice the health of future dwellers in 
 the houses so built for the sake of the small gain obtained 
 by the use of this filthy matter is simply unpardonable. 
 
 The volcanic substance known as " puzzuolana," which 
 is so common in Italy, has long been used with the best 
 effect. This substance derives its name from Pozzuoli, near 
 Naples, where it is found in great quantity. It is the basis 
 of the best Roman mortars, both ancient and modern. It 
 is shipped to England from Civita Vecchia. The colour of 
 the material is either grey, reddish brown, or violet. It is 
 roughly granular, resembling a cinder. 
 
 Similar deposits have been found in France, but the variety 
 best known in England is that found near Audernach, on the 
 Rhine, where it is known as tarrass or trass. One effect of 
 
23 
 
 it is to render rich limes highly hydraulic, as well as im- 
 proving the hydraulic limes. According to Vicat, mortars 
 made with puzzuolana begin to set under water the first day, 
 begin to gerhard on the third day, and are as hard as bricks 
 at the end of twelve months. The proportion in which it 
 is used by the French authorities is i of lime to z\ ot 
 puzzuolana, or 2 of trass ; or i of lime to i of sand and i 
 of puzzuolana. 
 
 This substance is really a silicate of alumina, with some 
 oxide of iron, and in some cases more or less lime. 
 Berthier made an analysis of a sample of trass, and of one 
 of puzzuolana, with the following results : 
 
 Silica 
 Alumina 
 
 Trass. 
 .. .. 57-0 .. 
 
 I2*O 
 
 Puzzuolana. 
 44'5 
 
 I c - O 
 
 Lime 
 
 .. .. 2-6 .. 
 
 .. 8-8 
 
 Magnesia 
 
 . . . . i *o 
 
 4-7 
 
 Oxide of iron 
 
 .. .. 5-0 .. 
 
 I2'0 
 
 Potash 
 
 7 " O 
 
 I " 4. 
 
 Soda 
 
 I'D . . 
 
 . . x T- 
 
 4'0 
 
 Water 
 
 Q'6 ' 
 
 Q " 2 
 
 
 
 . y -* 
 
 95 -2 99-6 
 
 Smeaton, after a long series of experiments with these 
 and other materials, employed them for mixing with lias 
 lime to make a cement for use in the construction of the 
 Eddystone Lighthouse. 
 
24 
 CHAPTER IV. 
 
 MORTAR AND CEMENT 
 
 HAVING considered the nature of the essential ingredients 
 of mortar and cement, viz., lime, silica, and clay, it is 
 necessary to consider the relation of these substances to each 
 other in the production of the resulting compounds known 
 as " mortar " and " cement." The use of cement in 
 connection with concrete will be considered later on. 
 
 The essential feature in the formation of both mortar and 
 cement is the hydration, or combination with water, of the 
 lime, known as "slaking," and the subsequent "setting" of 
 this material in contact with silica and clay. In the case of 
 mortar we must have a hydraulic lime containing, roughly, 
 about 3 parts of lime to i of clay. After slaking, this is 
 added to a certain proportion of silica in the shape of sand. 
 Other materials will answer to a certain extent, but, 
 undoubtedly, good, clean, sharp sand is the best. The 
 
proportion of lime to sand is generally very variable, and 
 the only legal definition of " mortar " is that contained in 
 the bye-laws of the London County Council, which specifies 
 that the mortar to be used under Section 16 of the 
 Metropolitan Management and Building Acts Amendment 
 Act, 1878, must be composed of freshly-burnt lime and 
 clean, sharp sand or grit, without earthy matter, in the pro- 
 portion of i of lime to 3 of sand or grit. In the case of 
 * l cement-mortar," the cement to be used must be Portland 
 cement, or other cement of equal quality, to be approved 
 by the surveyor, mixed with clean, sharp Fand or grit, in the 
 proportions of i of cement to 4 of sand or grit. Also that 
 burnt ballast or broken brick may be substituted for sand or 
 grit, provided such material be properly mixed with lime in 
 a mortar mill. 
 
 From the foregoing it will be seen that the composition 
 of lime-mortar answers to that of -Portland cement, when the 
 latter is gauged with sand ready for use. 
 
 In the manufacture of Portland cement, lime and clay 
 are mixed together, burnt, and ground into a powder. The 
 quantities of these ingredients are not quite the same as 
 those in hydraulic lime, the proportions being, roughly, 
 2 parts of lime (CaO) to i part of clay. The difference in 
 the resulting compound is mainly brought about by the 
 action of heat, by means of which the lime and silica of the 
 clay are caused to combine, with the result that when 
 moistened and allowed to set, the particular combinations 
 so feebly present in the mortar are present to a much larger 
 extent in the cement, with a consequent increase in its 
 strength. 
 
 It will thus be seen that the art of the cement manufac- 
 turer consists in the selection of materials and mixing 
 ihem together in such proportions as will produce a com- 
 
26 
 
 pound having in its finished state the maximum strength 
 obtainable. First, the quantity of lime (CaO) in the chalk 
 used is ascertained by chemical analysis, then the quantity 
 of carbonate of lime in the clay, if any, is determined. 
 The presence of silica, as sand, in the clay is also looked 
 for. Having ascertained these factors, the relative pro- 
 portions of chalk and clay required to produce a mixture 
 containing about 3 parts of chalk, or carbonate of lime if 
 a limestone is employed, to i part of clay, are calculated, 
 and a mixture is made by intimately grinding the two- 
 substances together. When this is done, a sample of the 
 mixture must be analysed to ascertain whether the desired 
 proportions have been obtained. If this be found not to 
 be the case, either more chalk, or clay, as the case may be r 
 is added, and the whole thoroughly incorporated and again 
 examined. Unless these precautions are observed, the 
 manufacture of Portland cement becomes a haphazard 
 proceeding, with great risk of the production of an inferior 
 article. 
 
 The desired mixture having been obtained, the mass is 
 dried on large floors heated by the waste heat from the kilns. 
 When thoroughly dry it is cut out in blocks and stacked 
 in the kilns, which in many cases take charges of 40 tons. 
 During the stacking fuel is mixed with the charge, and then 
 the whole is burnt until the mass is at a glowing heat. 
 Great care is taken that the mass is not overburnt, in which 
 case the material is spoilt. When properly done the 
 cooled masses have the appearance of a cinder. If the- 
 heating is carried to the extent of vitrification the cement 
 is useless. The burnt cement is then ground to a fine 
 powder, the strength of the cement depending very largely 
 upon the degree of fineness. This was not clearly under- 
 stood until the late Mr. John Grant, C.E., pointed oui the 
 
advantage of fine grinding. Since the publication of his-- 
 results the matter has received much attention, with the 
 result that greatly improved methods of grinding, such as 
 the roller mill, have been introduced, and a consequently 
 higher strength of cement obtained. 
 
 There is much difference of opinion as to the necessary 
 degree of fineness. Mr. Grant specified that the residue on 
 a sieve of 5800 meshes to the square inch (equal to about 
 76 per lineal inch) shall not exceed 15 per cent, by weight ;. 
 whilst Mr. Michell proposed only 10 per cent, on a 
 50 sieve, equal to 2500 per square inch. Messrs. Hilton,. 
 Anderson, and Brooks supply, when required, a cement 
 ground to loor 12 percent, residue on a mesh of 32, 000= 180 
 per lineal inch. 
 
 The further question of the specification for Portland 
 cement will be dealt with in the chapter relating to the 
 1 Technical Testing of Cement:" ' 
 
 Roman Cement. In 1791 Mr. James Parker took out a 
 patent for the manufacture of so-called " Roman Cement."" 
 This he prepared by burning and then grinding the ferru- 
 ginous nodules found in the Island of Sheppy. 
 
 This cement is used either alone or mixed with sand or 
 other materials, as desired for the particular purpose for 
 which it is to be employed. The following is the approxi- 
 mate composition of this material: Silica, 15 per cent.;, 
 alumina, 5 \ ; oxide of iron, 7 ; magnesia ; i : manganese, i ;. 
 lime, 45 ; moisture, carbonic acid, and loss, 26 per cent. 
 
 Similar cements to Parkers " Roman " cement are known 
 as " Atkinson's," made in a similar manner in Yorkshire :. 
 whilst Medina cement is made from nodules obtained in the 
 Isle of Wight. These were but the forerunners of Portland 
 cement. 
 
 Selenitic mortar was a device of Major General Scott, who- 
 
28 
 
 mixed with the cement a substance containing sulphuric 
 acid, namely, either plaster of Paris (sulphate of lime) or 
 green copperas (ferrous sulphate). This had the effect of 
 hastening the setting of the cement, but modern experience 
 is against this addition in all cases where great strength is 
 required. 
 
 Plaster of Paris. This valuable substance is a sulphate 
 of lime rendered anhydrous, or free from water, by heating. 
 When in a crystalline condition it is known as selenite, 
 largely used for the manufacture of polariscopic prisms, 
 whilst in pure milky-white masses it is known as alabaster. 
 The ordinary hydrated deposits, having the composition 
 CaSO.i 2 H..O, occur in various rock formations, and are 
 known us gypsum, which occurs largely in the proximity of 
 Paris, whence the name of the cement made from it. There 
 it is quarried very largely. The crude substance is heated 
 in kilns to a temperature of 240 deg. Fah., then finely 
 .powdered. The heating causes it to part with its water of 
 hydration, which begins to separate at about 175 deg. Fah. 
 When used, the dry plaster is mixed with water, which 
 rapidly re-combines with the anhydrous sulphate, and causes 
 it to set. 
 
 Keene's Cement. In consequence of the porosity and 
 -easily injured surface presented by plaster of Paris, many 
 attempts have been made to harden it. For this purpose 
 Keene heated the burnt gypsum with a solution of i part 
 of alum in 12 parts of water, at a temperature of 95 deg. 
 After standing in this solution for three hours, the gypsum 
 is dried, re-burnt, and ground to a powder. This plaster 
 is slower in setting than the ordinary plaster of Paris, but is 
 about half as hard again. 
 
 Parian Cement is plaster of Paris hardened with a i o per 
 cent, solution of borax. 
 
2 9 
 
 Stucco is plaster of Paris prepared with a solution of glue. 
 
 Lime water Has been recommended instead of fresh water 
 for mixing with plaster of Paris ; also a solution of gum 
 Arabic. Hardened plaster surfaces thus obtained may be 
 stained and polished as desired. 
 
CHAPTER V. 
 
 THE SETTING OF MORTARS AND CEMENTS. 
 
 As -already shown in the previous chapters, mortar is 
 essentially a mixture of hydrate of lime with three to four 
 times its weight of sand, and according to the text-books its 
 hardening is chiefly due to the gradual conversion of the 
 lime into carbonate by absorption of carbonic acid, known 
 strictly as carbon dioxide (CCX,), from the atmosphere. This 
 absorption is at first rapid, and then gradually slower. 
 .Mortars even centuries old are said to still contain appreci- 
 
.-able quantities of hydrate of lime. The hardening may be 
 in some cases partially due . to the formation of silicate of 
 calcium, but the evidence of this is not very strong. 
 
 It is generally held that pure silica in the form of sand 
 .acts merely mechanically, and enters into no chemical com- 
 bination with the lime. For all practical purposes this is 
 true, but experiments have shown that in the course of 
 several years some such action does take place to a very 
 slight degree. Petzholt found : 
 
 (1) That in mortar one hundred years old there was more 
 :soluble silica than in the original lime. 
 
 (2) That in mortar three hundred years old there was 
 three times as much soluble silica as in mortar one hundred 
 years old. 
 
 While these results may be accepted they yet leave the 
 -question open, as the slight action which had taken place 
 was too small to throw any light on the process of harden- 
 ing within the period of "setting." The degree to which 
 lime is carbonated is also so slight within reasonable limits 
 of time as not to warrant the assumption that the hardening 
 properties are due to this action. Doubtless the re-arrange- 
 ment of the molecules in what may be termed the initial 
 stage of crystallisation, plays an important part, and from 
 the fact that crystals of carbonate of lime adhere most 
 firmly to the sand, and will stand a considerable degree of 
 force before they can 'be released, it would appear that this 
 mechanical action forms one of the most important facts in 
 the problem. 
 
 It is well known that when lime is thrown out of 
 solution by the process of water softening, introduced by 
 the late Dr. Clark, the precipitated lime is generally in a 
 crystalline form, and it is not unreasonable to assume that 
 some such action takes place when mortar sets, the minute 
 
crystals formed adhering more or less firmly to the grains 
 of sand according to the conditions prevailing at the 
 moment. This action would go far to explain the cause of 
 the non-setting of mortar made with had or imperfectly 
 burnt lime, old mortar, &c. If the crystallisation of the 
 lime takes place upon the surface of a substance which in 
 itself is friable, the resulting mass will have a power of cc- 
 hesion greater than that possessed by the same substance 
 in proportion only to the power of the lime; just as the 
 strength of a brick wall cannot practically be greater than 
 the strength of the bricks used, as, however hard the 
 cement, the bricks will give way under a strain in excess of 
 the power of cohesion of their integral parts. Hence the 
 absolute necessity for the "grit" of mortar to be clean, free 
 from earthy matter, and of an unfriable nature. 
 
 This crystallisation theory also explains the reason why so- 
 many different substances may be used in making good 
 mortar a fact which the theory of chemical combination of 
 the component elements is altogether unable to reduce to> 
 order. 
 
 Another point is, that the modern practice of grinding all the 
 materials for making mortar together in a mill leads to a 
 most unsatisfactory condition of the "grit." The crystal- 
 lisation theory presupposes the presence of a hard and 
 fimly resisting number of particles cemented together by the 
 crystallisation of the lime, whether as carbonate, hydrate, &:c. 
 If, by grinding the grit to a powder, these " main supports," 
 as the particles of solid matter may be termed, of the 
 structure are broken down to a mere dust, then the initial 
 power of resistance to pressure which they exerted in their 
 unground state, will be thrown away, and a useless as well 
 as dangerous muddy mass be obtained in their place. 
 
 Marignac has observed that anhydrous calcium sulphate 
 
33 
 
 yields with water a supersaturated solution, which afterwards 
 deposits crystals of the hydrated salt. With plaster of 
 Paris heated at 140 deg., a solution can be obtained con- 
 taining 9 grammes of the salt per litre, or four times the 
 amount which exists in solution under normal conditions. 
 Le Chatelier* considers that the setting of plaster of Paris 
 is due to two distinct but simultaneous reactions. The 
 anhydrous calcium sulphate when moistened with water 
 dissolves and becomes hydrated, forming a supersaturated 
 solution, and this supersaturated solution deposits crystals 
 of hydrated calcium sulphate, which gradually increase in 
 volume, and unite one with another. This progressive cry- 
 stallisation continues so long as any anhydrous salt remains 
 undissolved. This theory is supported by the fact that in 
 practice 140 deg. is found to be the best temperature at 
 which to heat the plaster of Paris, and Marignac found 
 that the most highly supersaturated solutions are formed by 
 calcium sulphate heated to this temperature. It is found 
 that the addition of a small quantity of sulphuric acid or 
 sodium chloride to the water used for moistening the plaster 
 promotes setting. Both these compounds increase the 
 amount of calcium sulphate which can exist in the super- 
 saturated solution. 
 
 Similar phenomena of setting, due to deposition of crystals 
 from a supersaturated solution, are observed on moistening 
 coarsely - powdered anhydrous very soluble salts, such as. 
 sodium sulphate and carbonate, which readily form super- 
 saturated solutions. 
 
 This theory is applicable to the setting of all mortars^ 
 especially cements and hydraulic mortars. The solubility 
 of line is well known. Le Chatelier has recently shown 
 
 * Journal of Chemical Society, July, iSSj, p. 712. 
 
 D 
 
34 
 
 that calcium aluminate is soluble, and he hopes to prove 
 that calcium silicate is also soluble. 
 
 The character of the sand employed has also much to do 
 with the result. I made an experiment some years back in 
 conjunction with the late Mr. John Grant, whose work on 
 cements is so well known and justly appreciated, in which 
 two portions of a sample of good Portland cement were 
 made up into briquettes with a given sample of sand. In 
 the first the sand was in its normal state, but in the second 
 experiment I heated the sand to redness, so as to completely 
 oxidise the iron contained in it. The result was that the 
 breaking strain of the briquette made with the heated sand 
 was found to be nearly double that made with the unheated 
 sand. Possibly, the greater combining power may have 
 been due in some respects to an altered condition of the 
 face of the particles of sand, in consequence of which the 
 lime crystals adhered more firmly, and thus resisted the 
 strain to a greater extent. 
 
 In consequence of pressure of other work, the experi- 
 ments were not continued, but it may be that they would 
 form the starting point of a series which might result in 
 valuable information being obtained. 
 
 Now that means of obtaining excessive pressures are 
 more readily accessible than formerly, it might be of value 
 to undertake a series of experiments in which the briquettes 
 made of various limes, cements, and sands, should be 
 subjected to pressure of various degrees, and for various 
 periods, and their breaking strain ascertained. Such a 
 work might well be undertaken by a Committee of experts 
 having the proper facilities at their disposal, as it is obvious 
 the resources of a busy private laboratory are insufficient 
 for the purpose. 
 
 The recent experiments on the breaking strain of brick 
 
35 
 
 piers made with various cements, &c., by the Institute of 
 Eritish Architects might very well be extended in this 
 direction. In the course of a series of experiments under- 
 taken by a Committee of that body, a number of piers 
 constructed of various descriptions of 'bricks and cements 
 were kept for various periods, and then subjected to heavy 
 pressures, and the results have been published in the form 
 of two valuable reports, a review of which will be submitted 
 later on. 
 
 The late Mr. John Grant, in his account of his researches 
 on the strength of cements, presented to the Institute of 
 Civil Engineers in April, 1871, remarks "that the cohesive 
 varied directly as the adhesive strength of cement ; that, 
 in fact, neat cement was from three to four times the 
 adhesive strength of any mixture of it with four or five times 
 its bulk of sand ; and on this account he was prepared to 
 state in general terms that cohesive and adhesive strength 
 might be taken as equivalent to each other. For the 
 purpose of cementing together various materials, he need 
 hardly point out that to have strong cement or mortar was 
 not the only thing necessary. For instance, two cubes of 
 ; glass could not be cemented together, however strong the 
 cement, so that they should adhere with the same tenacity 
 .as two bricks or blocks of stone of a porous character and 
 rough on their faces. It would not be possible, for example, 
 to take such bricks as were recently used for lining the 
 .subway from the Houses of Parliament to the adjoining 
 railway station, and to cement their highly glazed china 
 faces together so that it would require the same pull to 
 separate them as it would to separate two stocks, wire-cut 
 gault, Fareham red, or other porous bricks, capable of 
 absorbing one-fifth of their weight [of water. Whilst on 
 this point, he mig'it state that on;y second in importance 
 
 D 2 
 
36 
 
 (if second) to the strength of the cement was that of seeing 
 that the bricks or stone to be cemented together were 
 thoroughly saturated with -water. In hot climates, like 
 that of India, any work being built with cement should be 
 kept wet and, if possible, under shelter during its progress,, 
 and for some time afterwards. Two bricks might be put 
 together dry, as they came from the kiln, with the strongest 
 cement that could be made, and if not previously soaked in 
 water these bricks might be separated without the least 
 difficulty, the joint being very little better than if made of 
 dry sand. With bricks which absorbed from i Ib. to 
 T % Ib. of water, it was evident that, if left to absorb this 
 moisture from the mortar with which they were cemented, 
 they took away that which was necessary for crystallising or 
 setting the cement." 
 
 The suggestion that the adhesive powers of mortar and 
 cement are due to a process of crystallisation in contact 
 with a roughened surface is strongly supported by the 
 results of Mr. Grant's experience, as he lays the greatest 
 stress upon the necessity for the presence of an ample 
 supply of still water. The experience with the dry brick is 
 most striking, as it is clear that such a brick would instantly 
 absorb the moisture in the cement immediately in contact 
 with it, and thus deprive it of the moisture necessary to- 
 bring about the process of crystallisation 
 
37 
 
 CHAPTER VI. 
 
 KOUGH TECHNICAL TESTS OF MORTAR WITHOUT CHEMICAL 
 
 ANALYSIS. 
 
 It is well known that in a large number of cases it is quite 
 possible for a man having experience in certain work to 
 detect almost at a glance the character of a material with 
 
 which he is daily work- 
 ing, and so it is with 
 mortar. To the practi- 
 cal eye there can be no 
 mariner of doubt as to 
 the kind of mixture 
 which is being employed 
 for the purpose of bind- 
 ing bricks. If it is good 
 stuff, made of well-burnt 
 lime and clean sand or 
 grit, the fact is patent at 
 once to the expert ; and, 
 so far as he is con- 
 cerned, there is no need 
 for a chemical analysis. 
 But when he has to 
 
 satisfy others of the correctness of his opinion then it 
 becomes necessary to resort to a method of registering in a 
 definite and special way the facts on which his opinion is based, 
 and this can only be done by having the sample in dispute 
 properly analysed, and its component parts weighed on the 
 balance, and the relative proportions of each material stated 
 
38 
 
 in precise figures. But if it is desired to place merely the 
 general facts before an independent person to convince him 
 that the mortar is not of a proper character, apart from the 
 question of the actual quantities of the component materials,, 
 then this can often be done by the adoption of the following 
 simple methods : 
 
 (1) Rough Test for Strength. Take a fragment of the 
 mortar between the thumb and forefinger, and try to break 
 it. If it crumbles easily into dust it is at once known that 
 that its power of cohesion is slight, and that it will be 
 wanting in binding power. 
 
 (2) Test for Vegetable Debris, &c. Mix some of the 
 mortar with an additional quantity of water, so as to make a 
 thin paste with it, and then stir it well with a large quantity 
 of water, adding the latter gradually, until the matters are 
 thoroughly disintegrated. If road scrapings, &c., are present,, 
 light organic debris will be seen floating upon the surface 
 of the water, and settling upon the top of the heavier mineral 
 matters when the whole is allowed to rest. 
 
 (3) Test for ascertaining the kind of Grit and relative pro- 
 portions of Lime to Grit. For this test a series of small- 
 glass cylinders with feet, so that they will stand upright on 
 a table, should be obtained. Their capacity should be about 
 100 cubic centimetres, or four ounces. Take about one 
 ounce of the mortar and well mix it with about an equal) 
 quantity of water, then introduce it into one of the cylinders,, 
 and add more water until the vessel is nearly full. Now 
 shake it thoroughly, having first closed the mouth of the 
 cylinder with a cork, so as to well diffuse the ingredients of 
 the mortar in the water, and thus separate the lime from the 
 grit as far as possible. This will soon be accomplished,, 
 when the cylinder should be allowed to rest. The com- 
 ponent parts of the mortar will now settle down in their 
 
39 
 
 order of gravity the grit first and then the earthy matter, 
 and finally the lime. . Of course, the line of demarcation 
 between the respective substances will not be very sharp, 
 but this will -not matter for our present purpose. If the grit 
 
 / 
 
 ROUGH TECHNICAL TESTS. 
 
 is composed of good clean sand it will be seen at once. In 
 the same way the presence of cinders, burnt ballast, &c., 
 will be instantly detected. If the lime be well burnt and 
 properly slaked, it will settle down last as a semi-flocculent 
 
40 
 
 < mass on the top of the grit ; but if it be old mortar or im- 
 perfectly burnt lime, then it will settle down more after the 
 manner of the grit itself, the difference between the beha- 
 - viour of the two, viz., the s;ood and bad lime, being most 
 marked. If the student will try the experiment on half-a- 
 dozen samples of mortar, some good and some bad, in as 
 many cylinders placed side by side, he will gain such an 
 insight into the method as can by no possibility be obtained 
 from a mere written description. 
 
 (4) Rough Test for Earthy Matter, e>r. This test is 
 very similar to the last, but with the addition of hydro- 
 chloric acid, known in commerce as " spirits of salt/' to 
 the mixture of mortar and water. For this purpose it is 
 better that the mixture should be made in a glass tumbler 
 or china basin, as in the event of a bad mortar much froth- 
 ing wi.l take place on the addition of the acid. If the 
 mortar is made with good, fresh, well-burnt lime very little 
 effervescence will occur, the lime dissolving in the acid 
 quietly ; but if it is made with old mortar or badly burnt 
 lime, then the acid will set free a large quantity of carbonic 
 acid, which, if there be a material quantity of earthy matter 
 or organic debris present, will cause a thick and dirty froth 
 to rise often to the rim of a 10 oz. beaker, when even only 
 a few grammes of the mortar are taken for the examination- 
 After the lime is all dissolved the earthy matters and grit 
 will be left, and on the mixture being poured into one of the 
 glass cylinders, and allowed to stand for a few minutes, it 
 will be seen at once by mere inspection whether the grit is 
 composed of proper or improper materials, such as road 
 detritus, ash-bin refuse, garden soil, &c. 
 
 (5) Tests for Setting Power. If a mortar is properly 
 made it will set under water, but if not it will remain in a 
 soft condition indefinitely. A convenient way to try this is 
 
4* 
 
 to place about i oz. of the mortar in question in one of the 
 cylinders above referred to, and cover it with water. In 
 ihis condition it should be allowed to rest for as long a time 
 as may be convenient, say a week or more. After the lapse 
 of sufficient time, if the mortar 'be pressed gently with a 
 glass rod, it will, if of good quality, be found to resist the 
 pressure, and be more or less hard ; but if it be composed 
 of rubbish, it will remain soft and allow the rod to be pressed 
 right down to the bottom of the cylinder with ease. 
 
 These tests are in themselves absolutely sufficient to war- 
 rant the condemnation, or passing, as the case may be, of a 
 given sample of mortar. But, unfortunately, they do not 
 enable the exact proportions of the respective ingredients to 
 be stated, and therefore, as I have said, chemical analysis 
 must be resorted to for this purpose. They are sufficient, 
 however, to enable one to show to independent persons, 
 such as members of committees, &c., the general character 
 of the substance which interested parties may be trying to 
 persuade them is "good mortar." 
 
 It sometimes occurs, unfortunately, that there is a dispute 
 as to the value of the indications of the results of an actual 
 analysis. In such cases physical tests are of the utmost 
 value. I was interested in a case recently in which I found 
 7 per cent, of " earthy matter," together with a large excess 
 of carbonate of lime, and accordingly reported that the 
 mortar was unfit for use. In consequence of representations 
 made by the builder, I presume, a further sample was sent 
 to another chemist, who evidently did not think that the 
 quantity of earthy matter was of importance, as he reported 
 in favour of the sample, with the result that the Local 
 Authority took no action. I afterwards learnt from the 
 surveyor who had charge of the case, that he had evidence 
 that earthy matter from under a hedge was used in place of 
 
42 
 
 clean grit. I think that there can be no doubt that if he 
 had made such simple tests as are here recommended, and 
 had placed them before his Committee, they would have 
 supported the first analysis and conclusions, authorised 
 action being taken to stop the use of the mortar, and 
 obtained an order on the builder to pull down the defective 
 brickwork. 
 
43 
 
 CHAPTER VII. 
 
 THE CHEMICAL ANALYSIS OF LIME MORTAR, CEMENT, ETC.,. 
 LIME AND LIMESTONE. 
 
 ALL samples should be first thoroughly broken up and 
 mixed so as to ensure the small portion taken for the actual 
 analysis being a fair representative sample of the bulk. 
 The quantity so prepared should at once be placed in a 
 wide-mouthed and well-stoppered bottle to preserve it from 
 atmospheric influences, otherwise it will absorb carbonic acid, 
 and possibly absorb or lose moisture, which would vitiate 
 the results. 
 
 Lime and Limestone may be considered as synonymous, 
 inasmuch as the lime is derived from the limestone by 
 calcination, the difference being due to the carbonic acid 
 which is driven out of the limestone in the process. 
 
 The chemical analysis of these substances may therefore 
 be conducted upon identical lines, as the same constituents, 
 have to be looked for in each case. These are : 
 
 Moisture. 
 
 Silica. 
 
 Oxide of Iron. 
 
 Alumina. 
 
 Lime. 
 
 Magnesia. 
 
 Alkalies. 
 
 Carbonic Acid. 
 
 Sulphuric Acid. 
 
 Several other substances are generally present, but these 
 may be overlooked in a technical analysis. 
 
44 
 
 For the estimation the following will be found to be a 
 suitable method of procedure : 
 
 (1) 5 grammes are taken for moisture. 
 
 (2) 5 grammes are taken for Silica, Iron, Alumina, 
 
 Lime, and Magnesia. 
 
 (3) 5 grammes are taken for Alkalies. 
 
 (4) i to 2 grammes are taken for Carbonic Acid. 
 
 (5) 5 grammes are taken for Sulphuric Acid. 
 
 (1) Moisture. 5 grammes of the sample are placed in a 
 platinum dish and dried at steam heat, or preferably in an 
 air bath at 103 deg. Cent, until it has a constant weight, 
 the loss representing the moisture contained in the sample. 
 
 Objection has been. taken to this method of ascertaining 
 the quantity of moisture present in the sample, in so far as it 
 may be assumed that carbonic acid will be absorbed by the 
 lime during the process of drying, but a large number of 
 experiments carried out by the author and Mr. R. Grimwood, 
 have shown that such absorption, if any, makes so little 
 difference that it is practically negligible. 
 
 (2) Silica, Iron, &c. It is generally preferable, in the 
 case of limestones especially, to place the 5 grammes taken 
 for the analysis in a muffle furnace, and heat to bright red- 
 ness for about an hour, by which means the carbonic acid is 
 eliminated, together with any organic matter that may be 
 present. By this precaution the excessive frothing which 
 would otherwise take place when the sample is dissolved in 
 acid is avoided, and consequently the loss from spurting. 
 
 The sample, when cool, is just moistened with strong 
 hydrochloric acid, and allowed to stand for fifteen minutes, 
 when practically the whole of the silica will be insoluble. 
 Water then is added, and the whole passed on to a filter paper, 
 xmd well washed till free from chlorides, and the filter paper, 
 containing the silica, &c., dried, ignited thoroughly, and 
 
45 
 
 weighed. When it is necessary to differentiate between sand 
 and combined silica, it .is necessary, after separating the 
 heavy sand by elutriation, to heat the ignited silica in a 
 platinum vessel* on the water bath with pure hydrofluoric 
 acid and a few drops of sulphuric acid, evaporate to- 
 dryness, ignite and weigh the residue. This may be 
 repeated until a constant weight is obtained, the total loss 
 by this treatment representing the combined silica. 
 
 Iron and Alumina. The filtrate from the above is 
 neutralised with ammonia, which is added in slight excess, 
 raised to the boiling point, and set aside for the iron and 
 alumina and settle out, or if the quantity be not large, they 
 can be filtered from the liquid at once. Having been 
 thoroughly washed with hot water, the precipitate is dried> 
 ignited and weighed as the mixed oxides of iron and 
 aluminium. 
 
 For the separation of these two constituents the weighed 
 precipitate is re-dissolved in hydrochloric acid, and treated 
 in either of the following ways : 
 
 a. Volumetrically by titration, with a standard solu- 
 tion of Bichromate of Potash, after reducing 
 the iron to the ferrous condition by means of 
 Stannous Chloride; 
 
 or b. Gravimetrically by separating the alumina from 
 the iron by means of a hot concentrated solu- 
 tion of caastic Potash.* 
 
 * The process is simple and reliable if the following directions are 
 adhered to. Dissolve the weighed oxide of iron and alumina in hydro- 
 chloric acid. This is facilitated by carefully grinding the gritty ignited 
 substance to a fine powder, placing it in a beaker, adding about 
 20 c.c. hydrochloric acid, placing the beaker on a wire gauze over a 
 Bunsen burner, and covering the beaker with a large watch-glass, 
 allowing the gas in the Bunsen burner to keep the acid simmering 
 gently, until the iron and alumina dissolve. When this is 
 
46 
 
 Lime. The filtrate from the original precipitation of the 
 mixed oxides of iron and aluminium is then treated with 
 oxalate of ammonium, taking care that there is a good excess 
 of ammonia present; boil briskly, remove the flame from the 
 beaker, and add gradually about 2 grammes of fine-powdered 
 pure solid ammonium oxalate while stirring continually. 
 Again heat to boiling. On removing the flame, the pre- 
 cipitate will settle rapidly, and is easily filtered and washed. 
 Having obtained the precipitate of oxalate of lime free from 
 oxalate of ammonium, the lime may be estimated in several 
 ways. 
 
 a. Volumetrically, by dissolving the precipitate in warm 
 dilute sulphuric acid, and titrating with standard permanga- 
 nate solution. 
 
 b. Gravimetrically, by the ignition of the dried precipitate 
 
 .accomplished, about 50 c.c. of a strong solution of caustic potash 
 should be placed in a beaker, and raised to boiling point over a 
 Bunsen burner. The next part of the operation requires 
 great care. If we add the whole of the hot acid, iron, 
 -and alumina solution at once to the hot caustic potash 
 solution, there. is a very great risk that the alumina will not 
 be all re-dissolved. The proper method is to add the acid solution 
 drop by drop to the potash, preferably letting it flow down 
 a glass rod, which must be used after every addition to thoroughly 
 stir the potash solution, and thus to ensure that the alumina, which 
 is at first precipitated by the potash, is thoroughly re-dissolved before 
 any further quantity of acid solution is added. From this it will be 
 seen how necessary it is to have an excess of the potash over the 
 quantity which would be necessary merely to neutralise the acid. 
 When the whole of the acid solution has thus been added to the 
 potash, the iron oxide will be in a state of suspension as a dark- 
 brown woolly mass. The solution should now be boiled for a 
 moment, and then filtered and well washed. As it is almost 
 impossible to wash out all the caustic potash, it is better, after 
 thoroughly washing, to re-dissolve the soft hydrated oxide of iron 
 
47 
 
 in a muffle and weighing as lime (CaO) ; or by its solution 
 in sulphuric acid, and precipitation therefrom as sulphate in 
 the presence of alcohol. 
 
 In exact work the oxalate of lime first obtained should, 
 after being thoroughly washed, be re-dissolved in HC1, and 
 re-precipitated, as it invariably carries down small quantities 
 of MgO. 
 
 The Magnesia is estimated in the filtrate from the lime 
 precipitate by evaporating the liquid to dryness with a few 
 drops of H..,SO 4J and gently heating to drive off ammonium ; 
 dissolve in about 50 c.c. of distilled water; add an excess 
 of ammonia and ammonium chloride, and then an excess of 
 a solution of phosphate of soda ; stir vigorously with a glass 
 rod, and put aside to settle for twelve hours. An alternative 
 method is to put the mixture of the solutions in a stoppered 
 cylinder, and at once shake vigorously for five minutes, when 
 
 on the filter, with a few drops of hydrochloric acid, first taking care 
 to place a clean beaker underneath the filter to collect the 
 chloride of iron thus formed. When all the iron is dis- 
 solved, wash all the ferric-chloride from the filter paper 
 into the beaker, add hot water, and then neutralise the acid with 
 ammonia as in the first instance, after the removal of the soluble 
 silica. This will precipitate the hydrated oxide of iron, which is 
 now of course free from alumina. On filtering the chloride of 
 potassium, formed by the action of the hydrochloric acid on the caustic 
 potash retained on the filte? and in the oxide of iron above described, 
 it will be easily washed away, when the precipitated iron oxide can 
 be dried, ignited, and weighed. This process is very much 
 neglected in these days of volumetric work, but it is one which I 
 have employed with great success. The whole secret is in the 
 patience of the operator, who must be careful to add the acid solution 
 very cautiously and slowly to the potash, which must be kept hot, 
 and stirred well after each addition, and in re-precipitating the 
 iron oxide with ammonia to get rid of the potash, which cannot be 
 washed out directly* 
 
4 8 
 
 the precipitation is completed without standing. Filter,, 
 dry, ignite, and weigh as magnesium pyro-phosphate, which, 
 on multiplication by 0*36, will give the weight of magnesia 
 (MgO). 
 
 (3) Alkalis. The most accurate method for the deter- 
 mination of the alkalis, soda and potash, is that of Professor 
 Lawrence Smith, which is fully described in Crookes' 
 " Select Methods in Chemical Analysis." The limestone or 
 lime is fused with one-quarter of its weight of ammonium 
 chloride. When cool, the crucible with its contents are placed 
 in a platinum dish, water added, and the whole slowly 
 heated. By this means the fused mass becomes detached 
 from the crucible and disintegrates. The crucible is then 
 washed into the dish, and the liquid filtered. The filtrate 
 from this contains the alkalis, together with some lime and 
 magnesia. The liquid is heated, ammonia and ammonium 
 carbonate are added in excess, and finally a few drops of 
 ammonium oxalate. When settled, the liquid is filtered,, 
 and the filtrate and washings concentrated until ammonium 
 chloride begins to crystallise out. The contents of the dish 
 are then transferred to a flask, and strong nitric acid is 
 added in sufficient quantity to decompose the ammonium 
 salts present. Heat, and, when the reaction is ended, transfer 
 to a porcelain dish, and evaporate to dryness. The residue 
 is then boiled with baryta water (barium hydroxide) to 
 remove magnesia, filtered, the excess of baryta being 
 removed by ammonia and ammonium carbonate. The 
 liquid from the last precipitation is filtered, and the filtrate 
 evaporated to dryness with a few drops of hydrochloric acid 
 in a platinum dish. The dish is carefully heated to ensure 
 dryness, and when cold the mixed chlorides of the alkalis 
 are quickly weighed. 
 
 An alternative method to the above is to dissolve the 
 
49 
 
 lime or limestone in hydrochloric acid, make the liquid 
 alkaline by the addition of an excess of milk of lime, and 
 heat and filter. It need hardly be mentioned that the 
 milk of lime here referred to must be made from freshly- 
 ignited pure lime, slaked, and made .into a cream with 
 distilled water. By this means all metals, except those of 
 the alkalis, are precipitated. The filtrate from thi 
 is then treated with ammonia, ammonium carbonate, and 
 ammonium oxalate, in the same manner as described in the 
 last method. 
 
 Separation of the Alkalis. The weighed mixed alkalis are 
 then dissolved in water, a few drops of hydrochloric 
 acid added, and finally, an excess of platinic chloride 
 solution. The dish is then heated on the water bath, until 
 the contents are in a semi-fluid condition. A little alcohol 
 is added, and the dish shaken to ensure mixture, when the 
 double salt of potassio-platinic chloride crystallises out, 
 and is filtered off, dried, and weighed. The weight so 
 formed multiplied by 0*153, will give the quantity of 
 potassium chloride, which, deducted from the mixed 
 alkaline chlorides originally found, gives the quantity of 
 sodium chloride. 
 
 Another method is to titrate the quantity of chlorine in 
 the weighed mixed chlorides, and calculate the result from 
 the following formulae : 
 
 We : ght of K = [(A- B) i -54] - B 
 0-63 
 
 Weight of Na = B - [ (A B) 0-91] 
 0-63 
 
 Where A = the weight of the mixed chlorides, and 
 B 2 the total weight of chlorine found in the chlorides, the 
 weight of each alkali is usually calculated as oxide. 
 
5 2 
 
 drying tube after the stopper on the end of the second tube 
 has been removed to allpw the access of air into the flask. 
 On cooling, the flask is again weighed, when the loss in 
 weight will represent the carbonic acid in the sample. 
 
 Numerous modiflcations of the above apparatus have 
 been made, and may be purchased at the instrument 
 
 C^ROTTER'S CARBONIC ACID APPARATUS. 
 
 makers. Perhaps one of the best of these is that devised 
 by Schrotter and shown in the above woodcut. It is used 
 in the same manner as the foregoing, except that concen- 
 trated sulphuric acid is used in the left-hand tube to dry the 
 carbonic acid. 
 
 (5) Sulphuric Add. The contents of the flask used for 
 the above determination of carbonic acid may now be washed 
 
53 
 
 out and the sulphuric acid in it estimated as follows, viz.: 
 Evaporate to dryness, gently ignite, moisten with hydro- 
 chloric acid, and. again dry, thus ensuring the complete 
 insolubility of the silica. Moisten again with hydrochloric 
 acid, add water, and filter, and to the clear filtrate add a 
 solution of barium chloride ; boil, and let stand to settle, 
 when the precipitated sulphate of barium is filtered off, 
 washed, dried, ignited, and weighed. The weight found, 
 multiplied by 0-3435, will give the quantity of anhy- 
 drous sulphuric acid (SO 3 ) present in combination as 
 sulphates. 
 
 MORTAR. 
 
 In the case of mortar, it is most important that a fair 
 average sample should be obtained. For this purpose the 
 preparation of the sample given under " Lime and Lime- 
 stone " should be carefully carried out. 
 
 A portion of the powdered sample, weighing 20 grammes, 
 is placed in a beaker, stirred up with a 10 per cent, solution 
 of hydrochloric acid, and allowed to stand for twenty minutes 
 in the cold, with frequent stirring. The fluid is then well 
 stirred, and allowed to stand one minute, when it is 
 decanted, and with it all the fine earthy matter, as well as 
 the organic particles, held in .suspension. This process is 
 repeated until the supernatant water at the end of one 
 minute is clear. The residue, which should be clean 
 washed sand and grit, &c., is then dried by washing it into 
 a platinum dish, draining off the water, drying the residue 
 on the water bath, and weighing the dish with its contents 
 from time to time until the weight remains constant. The 
 weight of the dish being then deducted, the remainder 
 represents the weight of "sand and grit" from the 10 
 grammes of the sample. 
 
6 
 
 fine matter insoluble in the weak solution of hydrochloric is 
 added to the sand and 'grit, and the above procedure 
 adopted. 
 
 CLAY. 
 
 Clay being a constituent of cement, it is necessary to 
 discuss this material in order to make the work complete. 
 
 As already pointed out in Chapter II. , there are several 
 varieties of clay, and for the purpose of analysis they may 
 be divided into two classes, viz.: 
 
 (1) Alluvial Muds and Clays free from Lime. 
 
 (2) Clay in which Carbonate of Lime is present. 
 
 In all cases it is desirable to work on the dried sample, 
 and for this purpose the clay should be thoroughly dried in 
 a water oven, when it should be at once ground into a 
 powder and placed in a stoppered bottle 
 
 (i) Alluvial Muds. Take 2 grammes of the dried 
 sample and transfer to a platinum dish. Add about 25 c.c. 
 of strong sulphuric acid, and dilute with 25 c.c. of distilled 
 water. Gently heat for several hours, taking care not to 
 raise it to the boiling point. Evaporate to dryness. When 
 cold extract with hydrochloric acid, which is first added in 
 sufficient quantity to just moisten the whole of the residue. 
 This is then put aside to stand for fifteen minutes, when water 
 is added, and the mass thoroughly disintegrated by means 
 of stirring, &c., with a glass rod. The sand and silica are 
 then filtered off, dried, and weighed. To separate the 
 " combined " silica from the sand, the weighed sand and 
 silica are repeatedly heated for some time (about 30 
 minutes) with successive portions of a strong solution of 
 sodium carbonate. By this means the silica formerly in 
 combination is dissolved, and the sand remaining behind is 
 then washed, dried, and weighed, the difference between the 
 
57 
 
 two weights giving the silica originally in combination with 
 bases in the clay. 
 
 The hydrochlo'ric acid solution is made up to a known 
 volume, and a part of it examined for iron, alumina, lime, 
 and magnesia ; a part for alkalies ; and another portion for 
 the estimation of sulphuric acid, as described above. 
 
 (2) Clay containing Carbonate of Lime. About 10 
 grammes of the clay are taken and treated with a slight 
 excess of dilute hydrochloric acid, filtered through a 
 weighed filter paper, washed, dried, and weighed, which 
 will give the weight of clay present. In the filtrate, iron, 
 alumina, lime, magnesia, &c., are estimated as under Lime. 
 The clay is then treated as above described for Alluvial 
 Clay. The whole of the results obtained are then calculated 
 to percentages on the original sample. 
 
 PORTLAND CEMENT. 
 
 Two grammes of cement should be treated with a small 
 quantity of water until all tendency to set has ceased. 
 Strong hydrochloric acid is then added, and the whole 
 evaporated to dry ness and heated to about i5odeg. C. It is 
 then moistened with strong hydrochloric acid, and again 
 dried, to thoroughly decompose the silicates. It is again 
 moistened with hydrochloric acid, water added, and the 
 whole thoroughly stirred up and filtered, washed, dried, and 
 weighed, the weight of the free silica plus sand, &c, being 
 thus obtained. The free silica is then estimated as described 
 under Alluvial Muds (see page 56). 
 
 The hydrochloric acid solution is now made up to a 
 known quantity, and portions examined, as already described 
 for iron, alumina, lime, magnesia, and sulphuric acid, under 
 Lime (see page 42). 
 
6o 
 
 and W. Fresenius have worked out 
 the following method of examina- 
 tion of Portland cement : 
 
 It is based on a comparison of 
 some of the properties and behaviour 
 under certain conditions of genuine 
 cement, and of substances used for 
 its adulteration, such as finely ground 
 slag and hydraulic lime. 
 
 For instance : 
 
 (a) Specific gravity. 
 
 (b) Loss by ignition. 
 
 (c) Behaviour to water; that 
 
 is, the alkalinity of the 
 aqueous solution. 
 
 (d) Behaviour to dilute acid. 
 
 (e) Behaviour to Chameleon 
 
 Solution (Permanganate 
 of Potash). 
 
 (/) Behaviour to Gaseous Car- 
 bonic Acid. 
 
 (a] The specific gravity is deter- 
 mined in the apparatus sketched on 
 this page, and is based on an accu- 
 rate determination of what space a 
 given weight of the cement occupies. 
 ] In making a determination, the 
 apparatus (see woodcut) is first 
 filled to the zero point with turpen- 
 tine ; 100 grammes of the cement is 
 then introduced with the aid of a 
 FRESENIUS' funnel into the apparatus through 
 
 SPECIFIC GRAVITY 
 
 BOTTLE. the graduated tube. The cement 
 
6r 
 
 at once falls to the bottom, and the liquid rises in the tube. 
 The number of c.c. read off on the tube represents the volume 
 occupied by the 100 grammes of cement. By dividing the 
 volume in c.c. into the weight, the specific gravity is obtained. 
 
 Precautions necessary in Manipulating. 
 
 Keep the lower portion of the apparatus in water at the air 
 temperature, and control by a thermometer, as constancy of 
 temperature is very important. 
 
 Give the apparatus a few gentle taps in order to facilitate 
 the escape of air bubbles. 
 
 Close the tube with a cork until complete subsidence has 
 taken place, in order to prevent evaporation of the liquid. 
 The results of two experiments ought not to differ more 
 than one unit in the second place of decimals. 
 
 The sample of cement in the above and for the following 
 experiments should be so finely pulverised as to pass 
 through a sieve with 5000 meshes to the square centi- 
 metre ; equal to 32,257 per square inch, or about 180 per 
 lineal inch. 
 
 (b) Loss by ignition is determined by heating 2 grammes 
 over an ordinary Bunsen lamp, with chimney, until weight 
 remains constant. 
 
 (c) Behaviour to Water -is determined by agitating i gramme 
 of finely powdered cement with 100 c.c. distilled water at 
 ordinary temperature for ten minutes. After filtration, 
 
 N 
 
 <$o c.c. are titrated with H Cl. 
 10 
 
 (d) Behaviour to Diluted Add is determined by agitating 
 i gramme of substance, finely powdered, for ten minutes 
 with a mixture of 30 c.c. normal acid and 70 c.c. water. 
 After filtration through a dry filter, 50 c.c. are titrated back 
 with normal soda solution, and from this the amount of 
 
6 4 
 0*5 grm. substance should correspond to 4 
 
 "N" 
 
 to 6 ' 2$ c.c. acid. 
 10 
 
 (d) The consumption of between i8'8 c.c. and 21-67 c - c 
 
 of normal acid by i grm. of cement. 
 
 (e) The consumption of between 0*79 and 2 '80 m.g, 
 
 of chameleon solution by i grm. cement. 
 (/) An 'absorption of from o to i '8 m.g. C O 2 by 3 grm, 
 
 cement. 
 
 Cements giving results which do not come within the 
 limits given must be looked upon either with suspicion, or 
 as beyond doubt adulterated, according to the results. 
 
 Table of Results of Experiments on Adulterated Cement. 
 
 
 
 
 i c 
 
 
 
 b 
 
 
 
 w 
 
 % -| 
 
 
 
 ti 
 
 u 
 
 
 >> 
 
 cT 
 
 3 V v 
 
 c3 
 
 S "Ja 
 
 
 Description and 
 
 rt 
 
 
 O w c 
 
 S8g 
 
 I'd 
 
 fl> c3 
 3 C 
 
 s 
 
 composition of the 
 mixture. 
 
 
 
 ~ o 
 >, w 
 
 |5S 
 
 *.& 
 
 " C 
 13 o5 
 
 a 
 
 1 
 
 
 o 
 <u 
 
 i " 
 
 |*j 
 
 g 
 
 s& 
 
 ctf 
 
 
 of 
 
 
 V 1 ^ ^frH *- 
 
 CJQ 
 
 ^.jT^^^ 
 
 Q 
 
 
 
 
 S c " ^ 
 
 ^^ 
 
 
 
 2 
 
 
 
 
 -o. 
 
 
 
 w 
 
 
 (a) 
 
 (*) 
 
 (0 
 
 (rf) 
 
 w 
 
 (/) 
 
 (i) Oneparthydrauliclime. 
 
 
 
 
 
 not 
 
 
 nine parts Portland 
 
 
 < 
 
 
 
 deter- 
 
 
 cement 
 
 3-067 
 
 1-90 
 
 6-50 
 
 20 -50 
 
 mined 
 
 4-6 
 
 (2) Oneparthydrauliclime 
 
 
 
 
 
 not 
 
 
 nine parts Portland 
 
 
 
 
 
 deter- 
 
 
 cement 
 
 ro53 
 
 2-52 
 
 8'20 
 
 20-04 
 
 miner 
 
 3-6 
 
 (3) One part of slag meal, 
 
 
 
 
 
 
 
 nine parts of Portland 
 
 
 
 
 
 
 
 cement 
 
 3-ii4 
 
 2-04 
 
 3-8 
 
 19-53 
 
 6-u 
 
 1-6 
 
 (4) One part of pulverised 
 
 
 
 
 
 
 
 slap, nine parts of 
 
 
 
 
 
 
 
 Portland cement 
 
 3-ii5 
 
 i-59 
 
 4-0 
 
 20'6 
 
 8-31 
 
 0-7 
 
65 
 
 The results under the headings a, t, d, &c., apply to 
 adulteration with slag : those under the headings a, b, c, and 
 / are of use in, the detection of adulteration with hydraulic 
 lime. 
 
 Interpretation of Table on page 64. 
 
 No. i shows too low specific gravity. Loss by ignition is 
 not normal. Alkalinity of aqueous solution too high, and 
 too great an absorption of CO 2 . 
 
 No. 2 shows too low specific gravity, a very high loss by 
 ignition, a high alkalinity of the aqueous solution, and too 
 great an absorption of CO 2 - In both cases sufficient proof 
 of adulteration by hydraulic lime. 
 
 Nos. 3 and 4. An admixture of slag meal can be recog- 
 nised by the large amount of cameleon solution used. In 
 No. 3 the alkalinity of the aqueous solution is below the 
 limit, while in No. 4 it is just reached. The specific gravity 
 of both is below that of genuine Portland cement, viz. 
 3-125. The consumption of acid gives no information. 
 
 On page 63 is a useful form of table for recording the 
 results of the chemical and physical examination of a 
 sample of cement. 
 
66 
 
 CHAPTER VIII. 
 
 TYPICAL ANALYSES OF MORTAR AND CEMENT. 
 
 DURING the past twenty years I have had opportunities 
 of analysing various samples of cement and mortar, 
 and of the sand which has been adopted as a standard for 
 use in testing the strength of cement having been associ- 
 ated with the late Mr. John Grant, engineer in charge 
 of the southern district of London and the building of the 
 southern portion of main drainage system under the late 
 Metropolitan Board of Works, whose work on cement has 
 for years been the standard of reference in regard to the 
 mechanical tests, &c. From my results obtained in the 
 course of this work, I have selected the following series of 
 typical analyses, and have added thereto the results of the 
 analyses of mortar, &c., which were first published in a paper 
 read before the Society of Public Analysts by myself and 
 Mr. R. Grimwood, F.I.C., &c., in 1896. As the value of 
 this paper is greatly enhanced by the discussion thereon, I 
 venture to print the whole in extenso, except that 
 portion devoted to analytical methods, which have 
 been already fully dealt with. Together these analyses 
 form a sequence which will be of interest and value 
 to those who are desirous of ascertaining the character ot 
 samples which have been submitted to analysis, as by com- 
 paring the results found with the following they will be able 
 to assign to any sample its proper position in the scale of 
 quality. Thus, it may be a mixture of good materials, or 
 of adulterated materials, or one which, though honestly made 
 up by the workman, has yet been mixed with poor mate- 
 
67 
 
 rials. I do not intend to imply, however, that the mere 
 analysis of a particular sample will indicate the setting 
 power of a cernent This, it will be seen on considera- 
 tion, must be affected more or less, if not entirely, by 
 the physical character of the cement, especially in regard 
 to the manner of its burning, grinding, &c. A cement may 
 contain all the ingredients of a first-class cement, but if it is 
 not properly prepared it will be useless. 
 
 THE LEGAL SPECIFICATION OF MORTAR. 
 
 In the bye-laws made by the London County Council, 
 under Section 16 of the Metropolitan Management and 
 Building Acts Amendment Act, 1878, it is provided that 
 the mortar used under that Act must be composed of freshly- 
 burnt lime and clean, sharp sand or grit, without earthy 
 matter, in the proportions of one of lime to three of sand or 
 grit. 
 
 In the case of cement-mortar, the cement to be used 
 must be Portland cement, or other cement of equal quality 
 to be approved by the District Surveyor, mixed with clean, 
 sharp sand or grit, in the proportions of one of cement to 
 four of sand or grit ; also that burnt ballast or broken brick 
 may be substituted for sand or grit, provided such material 
 be properly mixed with lime in a mortar mill. 
 
 In the bye-laws made by the Council under Section 31 
 of the London County Council (General Powers) Act, 1890, 
 it is provided that plastering or coarse stuff shall be com- 
 posed of lime and sand, in the proportion of one of lime to 
 three of sand, mixed with water and hair; but Portland, 
 Keene's, Parian, Selenitic, or other cement or plaster of 
 Paris, may also be used for plastering. The lime must be 
 freshly-burnt lime; the sand must be clean, sharp sand, 
 free from loam or earthy matter ; the hair must be good 
 
 F 2 
 
68 
 
 and sound, free from grease or dirt, and one pound of hair 
 be used to every three cubic feet of coarse stuff. Fibrous 
 material may be used instead of hair, and ground brick or 
 furnace slag, each to the satisfaction of the District Surveyor,, 
 may be used instead of sand ; and the setting coat must be 
 composed of lime or cement mixed with clean washed sand,, 
 or cement only. 
 
 USUAL CHARACTER OF MORTARS. 
 
 Earthy Matter. These extracts from the bye-laws 
 relating to the use of building materials in the metro- 
 polis, stating the exact proportions, enable an analysis 
 of lime or cement mortar, &c., to indicate very slight 
 departures from the prescribed quantities of the mate- 
 rials used. Unfortunately, analyses which have been 
 made from time to time by us (Dibdin and Grimwood)- 
 have shown distinctly that it is only under exceptional 
 circumstances that mortars which come within these 
 regulations are employed. To take one factor only, 
 the earthy matter. Analyses of samples of mortar, 
 made with materials selected as coming within the 
 definition of the bye-laws, and in the proportions therein 
 set out, show that the earthy matter, as determined by us 
 in the manner already described, varied as follows : 
 i'5, 2*1, 0*14, 074, o'Si, o - 6i, 0*64, 0*96, 0*85, and o'yo 
 per cent. ; thus showing that the finely-divided matters, 
 which may be accidentally introduced by means of the 
 broken brick or sand, form but a very small proportion of 
 the actual weight of the mortar. As compared with these,. 
 the following results obtained from the examination of a 
 number of samples of mortars taken from buildings in 
 course of erection will indicate the great difference in this 
 
THE 
 
 6 9 
 
 respect, viz. : 4-9, 3-8, 9-1, 3-4, 147, 137, 7-4, 10-4, 8'i, 
 1 2 -3 per cent. 
 
 Soluble Silica. Another factor of importance in esti- 
 mating approximately the quantity of cement in a genuine 
 cement mortar is the soluble silica. As is well known, the 
 silica present in Portland cement, soluble in hydrochloric 
 acid, will vary between 17 and 20 per cent.; but if the 
 average is taken at about 18 per cent., it will form a fair 
 basis for calculation. 
 
 If the cement mortar is made with good Portland cement, 
 and clean well-burnt brick or sand, in the proportion of one 
 of cement to four of the grit, there will be about 3 5 per 
 cent, of soluble silica. In some samples of genuine, but 
 badly mixed, cement mortar, the soluble silica was found to 
 vary from 2*0 to 3-25 percent.; and some samples of genuine 
 mortar made with lime gave soluble silica equal to 0-9, 
 i'o, 0*6, traces, and 1*1. If we now turn to the analyses 
 of commercial mortars (see Table V.) above referred to, 
 we find that the soluble silica varied as follows : 0-9, 0-7, 
 07; thus showing a marked difference in this factor in 
 good and bad mortars, &c. 
 
 Carbonic Add. A third factor is the quantity of carbonic 
 acid. Genuine mortars made with good materials were 
 found to contain, even after the mortar had been made for 
 some three months, only carbonic acid equal to i '02, 
 1-36, and i -06 per cent, of carbonate of lime; and 
 some cement-mortars 2-6, 2 '86, 1*04, 1*87, 1-45 
 per cent, of carbonate of lime. Freshly-prepared mortar 
 made with good materials contained in two instances o 3 
 and 0-4 per cent, carbonate of lime. Against these, some 
 commercial mortars contained 6-42, 6-24, 7 -39, 4*52, 8-9, 
 8-4, 8-0, and 4-5 per cent.; thus showing conclusively, that 
 this factor alone gave such marked indications of the character 
 
yo 
 
 of the material that it would be almost impossible that a 
 mistake could be made in estimating the character of a given 
 sample of mortar or cement. It should be noted that certain 
 bricks contain carbonate of lime, and this should be deter- 
 mined before concluding that the mortar is made with old 
 materials or insufficiently burnt limestone. In Table I. an 
 instance is given of a brick containing 7*77 per cent, of 
 carbonate of lime. 
 
 TABLE I. 
 Sample of Lime used in making Mortar. 
 
 Per cent. 
 
 Lime (CaO) 78-40 
 
 ,, carbonate 3*22 
 
 ,, sulphate 0-64 
 
 Soluble silica, iron oxide, and alumina.. .. 12-85 
 
 Sand 4-80 
 
 Total 99 -91 
 
 Sample of Brick used in making Mortar. 
 
 Per cent. 
 
 Moisture (loss at 212 deg. Fah.) 1-27 
 
 Lime (CaO) 1*99 
 
 carbonate 7-77 
 
 sulphate 1*03 
 
 Soluble silica, iron oxide, and alumina.. .. 3 '60 
 
 Insoluble matter coarse 78 77 
 
 Insoluble matter fine (this had the appearance 
 
 of having been burnt) 1-69 
 
 Total 96-12 
 
 Relation of Weight to Volume (Lime Mortar.} In. the 
 bye-laws of the London County Council it is provided that 
 the quantities are to be by volume, and not by weight ;. 
 therefore the analysis, being made by weight, must be con- 
 verted into terms of volume. In order to show how this is 
 
arrived at, the following comparative analysis of a sample 
 of mortar made with broken brick, specially prepared for 
 this purpose, will best make the point clear. The analyses of 
 the lime and broken brick are given in Table I., and in 
 Table II. that of the mortar made with one volume of lime 
 and three volumes of brick. In this case the brick had 
 some old mortar adhering to it, as would happen where old 
 bricks are used. 
 
 TABLE II. 
 
 Sample of Mortar made with One Part by Measure of Lime and Three 
 Parts Crushed Brick. 
 
 Per cent. 
 
 Moisture, water of hydration, &c 29-71 
 
 Lime (CaO) .. 12-85 
 
 (= 16-06 commercial lime.) 
 
 Lime, carbonate 4*47 
 
 sulphate o - 6i 
 
 Crushed brick 46 03 
 
 Iron oxide and alumina 2-80 
 
 Soluble silica i 85 
 
 Earthy matter (this had the appearance of 
 
 having been burnt) 0-70 
 
 Loss on ignition '. . .. 0-98 
 
 Total loo'oo 
 
 Commercial lime (containing 80 per cent, of CaO) to crushed brick, 
 i to 2-29 by volume ; to all other matters, i to 2*86 by volume 
 (see page 3). 
 
 In this it will be seen that the earthy matter was 070 per 
 cent, in the mortar, and the lime (as pure lime) 12*85 P er 
 cent , equal to 16*06 per cent, commercial lime of 80 per 
 cent, pure CaO. 
 
 The brick (Table I.), as already mentioned, contained 
 777 per cent, of carbonate of lime. Another instance is 
 given in Table III. of the analyses of lime and good brick, 
 and mortar made therewith. In this the carbonate of 
 lime was only 1-97 per cent, on the mortar. 
 
72 
 
 TABLE III 
 Analyses oj Stone-lime, and Mortar made therefrom 
 
 Mortar. 
 
 Stone Lime i : brick 
 lime. 3 parts jy 
 
 volume. 
 
 Moisture, water of hydration, and loss i'86 .. 31 -23 
 
 Lime (CaO) 78 -63 .. 
 
 Lime carbonate 0-93 .. 
 
 Lime sulphate None . . 
 
 Iron oxide and alumina 4*3 .. 
 
 Soluble silica yio .. 
 
 Sand and grit Trace . . 
 
 Earthy matter .. 6*20 .. 
 
 Loss on ignition .. 0*98 
 
 A large number of experiments on various samples of lime 
 have shown that it is a safe rule to increase the weight of 
 lime found in the analysis by one-fourth in order to raise the 
 lime (CaO) obtained to the relative quantity of commercial 
 lime, on the assumption that the commercial lime originally 
 contained 80 per cent, of pure oxide of calcium. The 
 correction from weight to volume is then made by means 
 of the relative specific gravities of the lime and sand or grit. 
 The following will illustrate the method : 
 Equal volumes of stone-lime and of air-dry broken brick 
 weighed as follows : 
 
 Lime 17*38 grammes per 18-5 c.c. 
 
 Brick 21-13 . > 
 
 (i) To convert the weight of commercial lime to a volume 
 cot responding to that of broken brick^ add one quarter of 
 the weight of lime found, to itself, thus: 17*38 + 4-34 = 
 21 72, as compared with 21-13 of brick. 
 
73 
 
 (2) By analysis this lime contained 78-63 per cent, of 
 real CaO, and a mortar made with this lime on analysis gave 
 10-89 P er cent - f CaO; therefore, to convert CaO in the 
 mortar into terms of comir.ercial lime, for practical purposes 
 add to the weight of CaO found one quarter of itself- 
 Thus, CaO in mortar = 10-89 F er Cnt - + 2 '7 2 = J3' 61 
 commercial lime. 
 
 (3) By No. i, 13-61 + 3-4 = 17-0 of CaO by 
 volume. 
 
 (4) The ground brick used in making the mortar (Table 
 III.) contained 6 per cent, of matters soluble in weak acid; 
 therefore, to the weight of brick-grit actually found add 
 6 per cent, i.e., sand and grit (brick), 45*99 (45*99 x ' c6 
 ~ 2-76) = 48-75. 
 
 (5) Lime to brick by volume = 17 : 49, or 1:3 
 nearly. 
 
 The following table of the weight of various sands and 
 other substances per cubic foot will be of assistance to the 
 reader. This table should not, however, be relied upon in 
 the case where a sample of the material under consideration 
 is obtainable ; in such a case it should be weighed under 
 normal working conditions. For comparison, I have also 
 given the weight of various limes, but in this case also it 
 will be best, wherever possible, to make a direct determi- 
 nation in each case. The weighings under the head of 
 " Sand, &c.," have been specially made for the purpose of 
 this chapter by Mr. C. Chambers Smith, C.E. 
 
 WEIGHT OF ONE CUBIC FOOT OF SAND, &c. 
 
 Ib. Ib. 
 
 Pi sand 100 Burnt clay 71 
 
 River sand 118 Cinders, clean .. .. 60 
 
 Broken brick . . . . 53 Clinker 50 
 
74 
 WEIGHT OF ONE CUBIC FOOT OF FRESH LIME. 
 
 In small lumps. Ib. 
 
 White chalk 36 
 
 Grey chalk, Hailing*.. 40 
 Portland stone . . . . 47 
 Blue lias . . .. 38 to 50 
 
 Ground. Ib, 
 
 White chalk 54 
 
 Grey chalk, Hailing*.. 58 
 
 Arden 68 
 
 Blue lias . . .. 49 to 68- 
 
 Flare-burnt grey lime. 
 CEMENTS. 
 
 
 Ib. 
 
 
 Ib, 
 
 Portland . . 
 
 . .74 to 101 
 
 Parian 
 
 .. 60 
 
 Roman . . 
 
 . .60 to 62^- 
 
 Plaster of Paris . . 
 
 .. 50 
 
 Medina . . 
 
 61 
 
 Whitening . . 
 
 .. 64 
 
 Keene's . . 
 
 64 1 
 
 
 RELATION OF WEIGHT TO VOLUME (CEMENT - MORTAR.) 
 
 Equal volumes of washed and dried sand and of Portland 
 cement of specific gravity 3*15 weighed 217 grammes and 
 19*3 grammes respectively. Therefore the weight of cement 
 found must be raised by one-eighth to bring the materials 
 into equal terms of volume. 
 
 As already stated, average Portland cement will contain 
 from 17 to 20 per cent, of soluble silica. An average 
 of 1 8 per cent, may therefore be taken, on which 
 assumption the soluble silica formed may be calculated 
 into terms of cement. Thus, the 3*25 per cent, of 
 soluble silica found in sample No. i, Table IV. y 
 would equal 18 per cent, of cement, or 19*25 per cent, 
 corrected for volume. The sand and grit equalled 67-34 
 per cent; therefore the ratio of cement to sand was 
 i : 3-5. In sample No. 2 of the same table the soluble 
 silica was 2-5 = 13-9 per cent, of cement = 15-6 by 
 volume; the sand, &c., was 7077 per cent.; the ratio con- 
 sequently was i : 4*5. Sample No. 3 showed a ratio of 
 i : 3-5, No. 4 of i : 5 '6, and No. 5 of i : 4-8, the average 
 
75 
 
 being i : 4*4. The samples were stated to be from 
 cement mortars, made with i of cement to 4 of sand,, 
 doubtless the differences are due to unequal mixing. 
 
 ABSORPTION OF CO 2 AFTER MORTAR IS MADE. 
 
 The crushed brick (see Table L), including some of the 
 silica, iron oxide, and alumina, would amount to 48 per 
 cent., or 5177 per cent., including 374 per cent, of 
 carbonate of lime found on analysis to be present in the 
 brick. The commercial lime used contained 3*22 per cent, 
 of carbonate of lime, so that the above 16*06 per cent, 
 would contain 0-52 per cent. These together will account 
 for 4-26 per cent, of the carbonate of lime in the quantity 
 found in the above analysis, showing that only 0*21 per 
 cent, of carbonate of lime was formed in the time between 
 the making of the mortar and its analysis. 
 
 The appended tables, Nos. IV. to VII., contain the 
 results of the analyses of some typical samples. 
 
 Table No. IV. shows the results of the analyses of five 
 samples of cement mortar. 
 
 Table No. V. shows the results of the analyses of three 
 samples of bad mortar. 
 
 Table No. VI. shows the results of partial analyses ot 
 fifteen samples of very bad mortar. In these the analyses 
 were carried only far enough to show that all the samples 
 contain excessive quantities of earthy matter and old 
 mortar, and also that they were deficient in freshly-burnt 
 lime. 
 
 In contrast with these, Table No. VII. shows the result 
 of the partial analysis of a sample of good mortar taken, 
 from a public building in course of erection. In all the 
 following tables the results are stated in percentages. 
 
7 6 
 
 TABLE IV. Analyses of Samples of Cement Mortar 
 
 
 No. i. 
 
 'NO. 2. 
 
 No. 3. 
 
 No. 4 . 
 
 No. 5. 
 
 Moisture, water of hy- 
 
 
 
 
 
 
 dration, &c. 
 
 7-50 
 
 6-12 
 
 7-72 
 
 7-26 
 
 9-68 
 
 Lime (CaO) . . . . 
 
 11-80 
 
 10-87 
 
 11-97 
 
 9-20 
 
 11-24 
 
 ,, carbonate 
 
 2-86 
 
 2-86 
 
 i -04 
 
 1-87 
 
 I- 45 
 
 sulphate 
 
 0-20 
 
 0-30 
 
 0-99 
 
 0-84 
 
 0-71 
 
 Iron oxide and alu- 
 
 
 
 
 
 
 mina 
 
 4'35 
 
 3-75 
 
 4-00 
 
 4'95 
 
 4-00 
 
 Soluble silica . . 
 
 3-25 
 
 2-50 
 
 3-25 
 
 2'OO 
 
 2-25 
 
 Earthy matter 
 
 0-61 
 
 0-64 
 
 0-96 
 
 0-85 
 
 o ' 70 
 
 Loss on ignition 
 
 2-09 
 
 2-19 
 
 2-42 
 
 2-50 
 
 2'02 
 
 Sand and grit 
 
 67-34 
 
 70-77 
 
 67-65 
 
 70'53 
 
 67'95 
 
 Total 
 
 lOO'OO 
 
 lOO'OO 
 
 lOO'OO 
 
 100-00 
 
 100-00 
 
 
 Ratio of Cement to Sand, &c. 
 
 
 No. i. 
 i : 3'5 
 
 Calculated on the Soluble Silica. 
 No, 2. No. 3. No. 4. 
 .. 1:4-5 i:3-5 1:5-6 
 
 No. 5. 
 .. i : 4-8 
 
 
 Calculated on the Lime. 
 
 
 i : 2-7 
 
 1:3-0 .. 1:2-9 I: 3' 
 
 .. 1:3-0 
 
 
 TABLE V. Bad Lime Mortars. 
 
 
 
 No. i. 
 
 No. 2. 
 
 No. 3. 
 
 Moisture, w T ater of hydration, &c. 
 Lime (CaO) 
 
 17-1 
 
 6-0 
 
 15-1 
 6-2 
 
 ,, carbonate 
 
 8-9 
 
 8-4 
 
 8-0 
 
 , sulphate 
 
 1-8 
 
 i -7 
 
 2 'O 
 
 Sand, grit, broken brick, &c. 
 Iron oxide and alumina 
 Soluble silica 
 
 5*'5 
 
 1-6 
 
 O'Q 
 
 46-3 
 0-7 
 
 53'4 
 
 Earthy matter 
 
 7 ' 4. 
 
 IO"4 
 
 8-1 
 
 Loss on ignition 
 
 4'9 
 
 4 -0 
 
 
 Total 
 
 IOO*O 
 
 lOO'O 
 
 lOO'O 
 
 
 
 
 
 Commercial lime (CaO) to sand and 
 grit, &c., by volume 
 Commercial lime (CaO) to all other 
 matters by volume 
 
 1 to 5-5 
 i to 8-3 
 
 i to 4-9 
 i to 7-7 
 
 i to 5-5 
 i to 8-1 
 
77 
 
 >-1 
 JS 
 
 t^ 
 
 
 ro 10 
 
 
 N t^ fOOO N N 
 
 10 00 
 
 K 
 
 roo O m O PO 
 
 M LT) 
 
 t-H t-f 
 
 
 
 O O O Tf O 
 00 ro CNO 
 
 t^ 
 
 LO o 
 
 
 rOO Th to O M 
 M 10 
 
 
 LO 
 
 6 
 
 
 C\ M O (^ <O O 
 
 r-> M t^ o O ro 
 
 NO iO CTv t^ ro 
 
 M IO 
 
 M LO 
 
 
 
 o 2 
 
 rt- 
 
 6 
 
 J-^O t^ 
 N C>. M OO t^ ^ 
 
 M 10 
 
 00 Tf 
 rf t^ 
 
 -S -2 
 
 ro 
 
 N OO IO Tj-CO O 
 ^}-O * M O O 
 
 * <* 
 o o 
 
 
 ro u-) iO t^CX3 ro 
 
 M IO 
 
 2 3 
 
 M 
 
 M N 00 >O O O 
 O t>- M 00 ro C^ 
 
 N HI 
 
 V 
 
 
 rODO PO "* t^ r< 
 
 M LO 
 
 -4-1 -*- 
 
 o 
 J? 
 
 rOOO M O TJ- 10 
 
 H TJ-O O Tf l>. 
 
 ro t^ o) 00 OO ro 
 
 M 
 
 
 
 10 ts 
 
 2 2 
 
 
 
 2:2: 
 
 
 
 .s 
 
 C "rt 
 
 
 O 
 
 0) . c . 
 C 
 J-l >- 
 
 
 o 
 
 &, '^3 * 
 
 o o 
 
 
 
 00 <D ' 
 
 
 1 
 
 -~ rt 
 .2 
 
 
 t3 
 
 U 
 
 %lo 
 
 
 
 r^ ^ rf : 
 
 
 Moisture 
 Lime (CaO) 
 Lime carbonate 
 Sand, grit, &c. . . 
 Karthy matter . . 
 Iron oxide, alumina, 
 
 Commercial lime (( 
 all other matters I 
 Commercial lime (C 
 by volume 
 
 10 
 
 6 
 55 
 
 N O IO LO ro O 
 ro C\ O< O N ro 
 
 O iO O N OO ro 
 IN 10 
 
 LO G\ 
 
 2 2 
 
 HI M 
 
 4- 
 
 6 
 
 iO -tf- O >O HI O 
 HI ro t^ ro Tt- N 
 
 rooo O O O rf 
 
 l-l LO HI 
 
 . 00 l^ 
 
 ro O 
 2 2 
 
 ro 
 
 t^ Ci ro t^ O O 
 C\ M O r-4 r^ t^- 
 
 ro 
 ro to 
 
 6 
 
 t^ O 00 t-oo ro 
 
 M LO 
 
 
 M 
 
 ro ro O C CO ' O 
 O- O 00 O t> C^ 
 
 <f IS 
 
 C 
 
 ro IX LO O 00 <* 
 
 
 M 
 
 OO HI O r> ro LO 
 00 N M 
 
 N Tt- 
 
 
 
 ro M OO <OOO -<j- 
 
 M LO 
 
 
 6 
 
 ^ HI LO C 
 O LO HI ro ro OO 
 
 ro 
 
 1 
 
 * r^ LO o o\ ro 
 
 l-l LO 
 
 2 2 
 
 l-l 1-H 
 
 o 
 
 c 
 
 00 O "OO OO O 
 
 N O H, ^ t^ O 
 
 o r> HI o oo ro 
 
 l-l l-l LO 
 
 O O 
 ro LO 
 
 2 2 
 
 00 
 
 t> r^ LO o o o 
 
 OO O N O M 10 
 
 O to 
 ro LO 
 
 z 
 
 ro M t^CO ^- 
 
 
 
 (U 
 
 OJ ^ ' |j ^ " 
 
 
 J 
 
 O o ,* 
 
 
 : : : : :o : 
 
 00- leg 52 I 
 o <u ."o 5 
 
 
 G ' 
 
 ri 
 
 H -Q rrt 
 
 
 Moisture 
 Lime (CaO) . . . 
 Lime carbonate 
 Sand, grit, &c. . . 
 Karthy matter . . . 
 Iron oxide, alumina, 
 silica 
 
 Commercial lime (C 
 cent, to sand, grit, 
 &c., by volume 
 Commercial lime (C 
 cent, to all other 
 volume 
 
73 
 
 The average total lime in cement may be taken as 60 per 
 cent; therefore, the lime found, including that carbonated, 
 may be corrected to volume of cement thus : Sample 
 No. i CaO = n -8 + i '6 (which had become carbonated) 
 =:i3'4 22'3 of cement ; add one-eighth for correction 
 to volume = 25-1 vols. of cement to 67-34 of sand, or a 
 ratio of i : 2 "j. 
 
 TABLE VII. Sample of Good Mortar taken from a Public Building 
 in Course of Erection. 
 
 Moisture I 7'84 
 
 Lime (CaO) I 4'43 
 
 Lime carbonate 0-97 
 
 Sand, grit, &c 59 -I 5 
 
 Earthy matter 1-05 
 
 Iron oxide, alumina, and soluble silica . . 3 -75 
 
 Commercial lime (of 80 per cent. CaO) to sand, grit, &c., 
 
 by volume, i to 2-6. 
 Commercial lime (of 80 per cent. CaO) to all other matters 
 
 (dry) by volume, i to 3-0. 
 
 It will be noticed, however, that the calculation of cement 
 from the lime found is unsafe, as the mortar might be 
 sophisticated with lime, and therefore not cement mortar. 
 The soluble silica is the only reliable indicator. 
 
 In opening the discussion which followed the reading of 
 the paper, from which the above is extracted, 
 
 Sir Charles A. Cameron said he was afraid he could not 
 add anything to the very valuable information that had 
 been laid before the Society by the authors of this paper. 
 He occasionally had specimens of cement submitted to him 
 which, although apparently having a correct composition 
 according to what was laid down in the text-books never- 
 theless would not adhere to the surface of bricks. As a 
 matter of fact, disputes were constantly arising in connec- 
 tion with cement, mortar, and concrete, and there was a 
 very large field open to workers in the direction of obtain- 
 
79 
 
 ing information which would enable such disputes to be 
 settled on a scientific basis. In Ireland he did not think it 
 was customary to use ground bricks in place of sand to 
 such an extent as appeared to be the case in England. - He 
 believed the authors had mentioned the case of a mortar in 
 which the proportion of lime to sand was as low as 1:9. 
 He had himself once examined a sample containing almost 
 as much sand, viz., 86 per cent. 
 
 Mr. Allen said that one problem which he had been 
 called upon to solve was how much road-sweepings a parti- 
 cular sample of mortar contained, and what was the 
 proportion of "clean, sharp sand." Road-sweepings seemed 
 to be a very indefinite article, and he thought that if the 
 authors could indicate how to ascertain the quantity present, 
 it would be of great service to others as well as to himself. 
 It seemed to him, if he had understood that part of the 
 paper correctly, that it was a little questionable whether it 
 was fair to deduce the amount of cement in a cement 
 mortar from the amount of soluble silica. This surely must 
 be a very variable quantity. There was no doubt that the 
 Society was very much indebted to anyone bringing forward 
 figures* on such a difficult subject, as they would at least 
 afford something like a basis for further investigations. 
 
 Mr. Bodmer remarked that where broken bricks were 
 used in substitution for sand, the bye-laws, although pro- 
 viding for the proper grinding up of the bricks, did not 
 specify distinctly that they should be well burnt. Badly- 
 burnt bricks might be used, and in that case the amount of 
 earthy matter present would be considerable. He thought 
 it desirable that either the use of bricks should be altogether 
 prohibited, or that some more definite limit should be laid 
 down as to their employment. 
 
 The proportion of quicklime contained in the lime used 
 
8o 
 
 was also a matter upon which disputes often arose, but it was 
 hardly possible to suggest any definite standard, for in some 
 kinds of lime, which were really very good for mortar 
 making, the percentage of actual quicklime was compara- 
 tively low. 
 
 Mr. John Hughes said that the bye-laws of the London 
 County Council in regard to the composition of mortar 
 were far too vague. Nothing is mentioned respecting the 
 quality of the lime, except that it should be "freshly burnt,' r 
 its richness in CaO and soluble silica is entirely omitted, as 
 well as its weight per cubic foot, which may vary consider- 
 ably. The sand must be sharp and clean, " without earthy 
 matter," but there is no definition of what constitutes earthy 
 matter. 
 
 In making mortar it is usual to specify quantities by 
 measure rather than by weight, consequently the respective 
 amounts of lime and sand, or substitutes for sand, must 
 vary in proportion to their gravity. Therefore mortar, 
 though made according to the strict reading of the bye-laws, 
 may contain as little as 10 parts of lime (CaO) by weight in 
 every 100 parts of the perfectly dry mortar. Thus, taking 
 a cubic foot of grey lime containing 80 per cent. CaO, and 
 weighing (according to Hurst's "Architectural Hand-book") 
 44 lb., and three cubic feet of clean sand weighing loolb. 
 per cubic foot, we have a mixture in the dry state which 
 contains only 10-23 parts of CaO in every 100 parts by weight. 
 On the other hand, if a substitute for sand, such as broken 
 brick, having a less gravity, be used, the percentage of CaO 
 in the mortar would be increased. Some years ago he de- 
 voted considerable time to the examination of old mortar 
 obtained from numerous old abbeys and castles, and the 
 results of the analyses were published in The Builder during 
 the years 1892 and 1893. These analyses corroborated 
 
8i 
 
 Mr. Bodmer's observation, that the percentage of actual 
 quicklime was really no criterion of the quality of the 
 mortar. The Ihne rapidly becomes converted into carbon- 
 ate, especially in the smoky atmosphere of towns, and in the 
 case of good building lime the CaO is associated with 
 soluble silica in the same form as it exists in Portland 
 cement, so that the total lime in its various combinations, 
 rather than the caustic lime alone, should be considered 
 before condemning a mortar. 
 
 He would mention that, according to his researches, the 
 average amount of caustic lime in ancient mortar did not 
 exceed 0-5 per cent. Further, within certain limits the 
 actual percentage of total lime was no reliable indication of 
 the quality of the mortar. Thus the mortar from Rochester 
 Castle (of which only the keep remains) contains as much 
 as 28-67 CaO, while that from Tintern Abbey, in Mon- 
 mouthshire (which was remarkable from the fact that the 
 four gable ends were still standing), contains only iS'S4 
 CaO per cent. The mortar from the leaning tower of Caer- 
 philly Castle, in Glamorganshire, contains 'as little as 13 -49, 
 associated, however, with 9-85 of soluble silica. He could 
 not agree with Mr. Dibdin in calculating the probable 
 amount of cement present in a mortar from the figures for 
 soluble silica. Limestones contain varying quantities of 
 gelatinous or hydrated silica, which after calcination forms 
 a natural cement, and it is to the presence of this cement 
 that the durability of ancient mortar largely depends. 
 Greystone lime contains 9 per cent, of this soluble silica ;. 
 Aberthaw lime contains as much as 15, while the best Port- 
 land cement contains not more than from 20 to 22 ; conse- 
 quently limestones are suitable for making mortar in 
 proportion to their richness in this so-called soluble silica. 
 In conclusion, he would say that some of the best and most 
 
 G 
 
82 
 
 durable of these old mortars contained as much as 4 to 5 
 per cent, of oxide of iron and alumina. 
 
 Mr. B. E. R. Newlands said he presumed this paper was 
 intended to refer only to cases occurring in those districts 
 in which the bye-laws referred to had been adopted ; but 
 there were districts where the bye-laws had not been 
 adopted ; and a still greater number where they had not 
 been enforced. In Manchester, and other large towns, 
 mortar was manufactured as an industry by the Corporation, 
 and the mortar was generally made with the cinders and 
 clinkers from the Corporation's destructors. This mortar 
 which was sold to the general public by the Authorities 
 would yield very different analytical results from the 
 mortars dealt with in this paper. Mortar made with furnace 
 clinker was also used to a considerable extent in the metro- 
 politan area most frequently, perhaps, in the case of build- 
 ings erected by works proprietors whose furnaces supplied 
 the clinker ; and this mortar, although of a very high 
 standard as far as hardness and setting power were 
 concerned, would not satisfy the tests given in the 
 paper. 
 
 Dr. M. A. Adams said that, from the local experience he 
 had gained in the midst of the cement district, where there 
 was an abundance of materials for making lime from both 
 while and grey chalk, he could confirm the remark made by 
 Mr. Hughes, that alumina was often a somewhat important 
 constituent of commercial lime. It was customary in his 
 neighbourhood to look upon the lime made from grey chalk 
 (which contained a large proportion of silicates and 3 or 4 
 per cent, of alumina) as almost equal in quality to Portland 
 cement. Certainly, it was a fact well known to all practical 
 builders in the district, that they could use far less of such 
 lime than was required if they employed lime made from 
 
83 
 
 the upper layers of the chalk, which was nearly pure calcium 
 carbonate. 
 
 The author, in replying to the discussion, said that he and 
 his colleague congratulated themselves on having raised such 
 an interesting discussion. He sincerely hoped that the paper 
 would be the forerunner of several others on the same and 
 kindred subjects. Mr. Newlands was perfectly correct in his 
 remarks as to the scope of the paper. The question had been 
 clealt with entirely as under the bye-laws relating to London. 
 As a matter of fact, these bye-laws contained the only state- 
 ment or definition of mortar having legal weight of which he 
 was aware. If they were wrong, it was without doubt desirable 
 i hat they should be amended as soon as possible, but as 
 they existed, a considerable amount of work had been done 
 under them, and it was just as well that they should be 
 thoroughly understood. The subject, taken in its industrial 
 .aspect and as ranging over the ground of all kinds of mortar, 
 both ancient and modern, was of course a very large one, 
 and to treat it exhaustively would require a great deal more 
 than the hour or so available at one of the Society's meet- 
 ings, artd he and his colleague looked upon this paper rather 
 as an index for pointing out that a large amount of valuable 
 work could be done by those who took up the subject in 
 detail. With regard to the adulteration of mortar, it was 
 seldom that a sample was found which really complied with 
 the specification contained in the bye-laws, a common 
 .adulterant, in London at any rate, being dust-bin refuse. 
 Mr. Allen had asked for information as to the best mode of 
 estimating road-sweepings. In many cases road-sweepings 
 consisted largely of sharp sand mixed with organic matter. 
 The only way to deal with these was to separate the organic 
 from the mineral m?tter, and estimate them in the ordinary 
 Of course, quite a cursory observation would indicate 
 
 G 2 
 
34 
 
 whether the organic matter was in the form of horse-dung 
 or the like. With regard to Mr. Bodmer's remarks, he did 
 not think much difficulty would be found in differentiating 
 between properly-burnt earthy matter and clay that had 
 merely been warmed up. The bye-laws appeared to be 
 perfectly definite in their references to the ballast or bricks- 
 that might be used in making mortar. They stated that 
 burnt ballast or broken brick might be substituted for sand 
 or grit. Now, earthy matter that had merely been warmed 
 could not properly be described as "burnt ballast," and. 
 unless a brick has been fired and properly fired it could 
 not, according to his (Mr. Dibdin's) idea, be regarded as- 
 coming within the meaning of the term " brick," any more 
 than partially heated chalk could be regarded as lime. Mr. 
 Hughes had remarked that the amount of lime present was 
 no criterion of the quality of a mortar, and had referred 
 to the mortar of Tintern Abbey as containing 18*84 per 
 cent, of lime. Now, assuming 18*84 per cent, of caustic 
 lime to be equivalent to 23*55 P er cent, of commercial lime, 
 and adding one-fourth to correct for the difference in specific 
 gravity between the lime and grit, the volumetric proportion, 
 of lime to grit worked out in this case to i 12-4, or more 
 lime than that laid down in the bye-laws. With regard to 
 clinker mortars, these were not comprised in the present 
 paper, which dealt only with good brick, grit, or clean, sharp 
 sand, and lime or Portland cement. In the case of mortar 
 made with other materials, other data must be obtained as- 
 a basis of calculation. 
 
 The presence of soluble silica in cement mortar must not 
 be taken as indicating the actual presence of Portland 
 cement, but that the quantity of Portland cement, if any, 
 could not exceed that indicated by the silica. If there is- 
 no soluble silica, no Portland cement could be present. 
 
CHAPTER IX. 
 
 MECHANICAL TESTS OF CEMENT. 
 
 Specification for Portland Cement, &c. The following 
 specification for Portland Cement and for Cement Mortar 
 was prepared by Mr. Frederick H. Lewis, and submitted 
 
 by him for the purpose 
 
 of discussion at the 
 Engineers' Club of 
 Philadelphia in 1894.* 
 
 ON THE MEDWAY. 
 
 GENERAL. 
 
 (i) Records and Sam- 
 ples. Parties offering 
 cements for engineering 
 works must be prepared 
 to file satisfactory re- 
 cords of tests, both neat 
 and in sand mortars, for 
 periods up to at least one year, and also records of analyses, 
 showing proportions conforming to recognised standards, 
 and to the chemical requirements below. Such tests and 
 analyses must be made by competent parties, certified by 
 them to be correct, and, if required, sworn to. 
 
 Prior to the award of contracts, parties wishing to be 
 considered will submit samples, duly marked for identifica- 
 tion, and each guaranteed to be an average sample of the 
 cement to be furnished in the event of an award. The 
 sample will be used for preliminary tests, and also preserved 
 for comparison with the cement delivered for use. 
 
 * " Proceedings" Eng. Club, Phila., 1895, xi., p. 325. 
 
86 
 
 (2) Testing. The briquette moulds, sand sieves, setting" 
 needles, &c., and the methods of making briquettes and of 
 testing, will all be in accordance with the recommendations- 
 of the American Society of Civil Engineers, with this 
 exception, viz., that the cement may be screened through a 
 No. 30 sieve, if necessary, to remove lumps and foreign 
 matter, but other than this, briquettes will be made up from 
 cements as samples without bifting. 
 
 Tests for setting lime, fineness, and the seven-day tensile- 
 tests required below, will be made from each sample taken. 
 The chemical tests will be furnished ly manufacturers- 
 before the award of contracts, but will be confirmed by 
 occasional tests of the cement furnished. 
 
 The requirements in tensile tests are for an average of 
 a set of five briquettes in each case. 
 
 (3) Fineness. Cements will not be considered unless- 
 they are ground so finely that 97-^ per cent, will pass, 
 through the standard No. 50 sieve, and SyJ per cent, 
 through the No. 100 sieve. 
 
 (4) Specific Gravity. Cements will be required to have 
 a specific gravity exceeding 3*00. 
 
 (5) Constancy of Volume. Mean made neat in pats, 3^ in, 
 in diameter, fin. thick at the centre, and drawn to a sharp 
 edge at the circumference, the cements must show no evi- 
 dence of cracking, checking, or warping when exposed in 
 the air, or in water at normal temperature, for one week. 
 
 (6) Sampling, Storing, and Accounting. All tests will 
 be made from average samples selected at random from 
 10 per cent, of the packages offered. Not more than 150 
 barrels will be accepted on one set of tests, and no cement 
 will be used which has not been duly sampled and tesu d 
 at seven days. 
 
 Contractors will provide good and sufficient storage fcr 
 
87 
 
 cements, so that there shall always be an ample stock on 
 hand, and will protect the stock against intrusion, or the 
 use of lots not authorised. 
 
 The Inspector will have free access to the storehouse 
 at all times, and will be afforded every facility for identi- 
 fying and accounting for different lots of cement in store. 
 He will be furnished promptly with duplicate invoices and 
 bills of lading of all cement for the work as shipments 
 are made. 
 
 In case of rejection, the lot of cement condemned must 
 be removed from store within twenty-four hours. 
 
 (7) Cost of Testing. Parties making propositions for 
 furnishing cements will include in their prices a sum 
 sufficient to cover the cost of having the cement regularly 
 sampled and tested in a manner satisfactory to the 
 purchaser. 
 
 (8) Limits of Accuracy. Chemical determinations will 
 be considered accurate to the nearest tenth of a per cent.; 
 tensile tests to the nearest ten pounds. Within these limits 
 tests must meet the requirements of specifications, or the 
 cement will be rejected. There will be no re-tests. 
 
 (9) ^Cements for Salt Water Exposures. Initial set with 
 fresh water, not less than ten minutes. 
 
 CHEMICAL. 
 
 Sulphuric acid (SO 3 ) less than i 'oo per cent. 
 
 Magnesia (MgO) less than i'oo ,, 
 
 TENSILE TESTS. 
 
 i day neat not more than 45 per cent. 
 
 of tests at 7 days ne^t 
 
 7 day neat .. 375 Ib. 
 
 7 days 3 parts sand to i of 
 
 cement 135 Ib 
 
 ( i o) Quick-setting or Untreated Cements for Fresh water 
 
88 
 
 Exposures. Initial set, with fresh water, not less than ten 
 minutes. 
 
 CHEMICAL 
 
 Sulphuric acid less than 1-25 per cent. 
 
 Magnesia, less than 2'5O ,, 
 
 TENSILE TESTS, 
 
 i day neat not more than 45 per cent. 
 
 of tests at 7 days neat 
 
 7 days neat 375 Ib. 
 
 7 days 3 parts sand to i 
 
 cement 150 Ib. 
 
 (n) Slow-setting or Treated Cements for Fresh-water 
 Exposures. Initial set not less than ten minutes. 
 
 CHEMICAL. 
 
 Sulphuric acid less than 2-00 per cent. 
 
 Magnesia less than 2-50 ,, 
 
 TENSILE TESTS. 
 
 i day neat not more than 45 per cent. 
 
 of tssts at 7 days neat 
 
 7 days neat 475 Ib. 
 
 7 days 3 part sand to i 
 
 cement 175 Ib. 
 
 MORTARS. 
 
 (12) Sand. The sand used shall be of such quality that 
 it will give in laboratory tests, 3 to i at 7 days, 60 per cent, 
 of the tensile strength required for briquettes made with 
 standard sand, and briquettes made directly from the mortar 
 box will similarly be expected to show 33 per cent, of the 
 test requirements for standard sand. 
 
 In laboratory tests of sand the lumps will be removed by 
 sifting through a No. 20 sieve. 
 
 (13) Mixing Mortars. The sand must in all cases be 
 
89 
 
 dry, and will be stored in such a way that it can be kept 
 dry. For mortars, sands will be sifted so as to exclude 
 lumps or particles larger than ^in. diameter ; for concrete, 
 to exclude those larger than %jn. diameter. In propor- 
 tioning mortars, the cements will invariably be measured 
 by weight ; the sand by bulk, to be agreed upon. A 
 thorough mixture of sand and cement will be made dry, 
 .after which the water will be added by sprinkling, and the 
 whole mass worked to a temper. 
 
 The amount of mortar to be mixed at one time will be 
 fixed to suit the character of the cements. No mortar will 
 be used which has been mixed over night, or over a length 
 of time fixed by the inspector. 
 
 (14) Lime in Cement Mortars. In the case of a mixture 
 of lime and cement in mortars, the lime must be slaked 
 at least ten days before using. 
 
 (15) Concrete. In making concretes, the cement and 
 sand will first be mixed dry in the manner set forth above, 
 and worked both dry and wet to a temper. The stones 
 measured by bulk will first be sprinkled and then added to 
 the mor&r and well worked in. 
 
 The concretes will be made with a minimum of water, to 
 .admit of being rammed in place without showing an excess 
 ,of surface water. 
 
 fineness. In the course of the discussion which followed 
 the reading of Mr. Lewis' paper, Mr. Robert M. Lesley 
 pointed out that the following finenesses are required by 
 the Governments of Europe : 
 
 Belgium 15 per cent, on a 5800 mesh sieve (Genie 
 
 Beige) 
 Germany 10 per cent, on a 5800 mesh sieve (Rule 
 
 Prussian) 
 Switzerland .. .. 15 per cent, on a 5800 mesh sieve (Report 
 
 Tetmajer, Zurich) 
 
9 
 
 France . . ... . . Xo fineness specification (Cahier des 
 
 Charges, Fonts et Chaussees) 
 
 Francs 5 per cent, on a 5800 mesh sieve (recom- 
 mended by Le Chatelier and Caudlot, 
 French Chemists) 
 
 Rouman ! a 12 per cent on a 5800 mesh sieve (Govern- 
 ment specification) 
 
 England . . . . . . 8 per cent, on a 2500 mesh sieve (recom- 
 mended by Faija, World's Fair Congress, 
 1893) 
 
 United States .. .. Recommendation of American Society of 
 Civil Engineers, June 2yth, 1893, says, 
 " Cements of better grades are ground 
 so fine that only from 5 per cent, to 10 
 per cent, is rejected on a sieve of 2500 
 meshes per square inch." 
 
 The following results of a series of tests of mortar made 
 with cements of increasing fineness, conducted by Messrs. 
 Colson and Colson * will be of interest : 
 
 Relation of the Fineness of Cement to the Tensile Strength of Mortar in Ibs. 
 
 Size of Sand. 
 
 Fineness of Cement used. 
 
 As received. 
 
 Through 
 2500 meshes 
 per sq. in. 
 
 Through 
 3600 meshes 
 per sq. in. 
 
 Through 
 5800 meshes 
 per sq. in. 
 
 Screened through 
 2500 meshes per 
 square inch . . 
 
 Ib. 
 
 122 
 
 Ib. 
 133 
 
 Ib. 
 203 
 
 Ib. 
 240 
 
 Each result is the average of six experiments. 
 weight of cement used in all the experiments. 
 
 Same 
 
 " Proceedings" Inst. C.E., Vol. xcv., Part I. 
 
Tensile Strength. Mr. Lesley quoted the following 
 specifications : 
 
 Neat Cements. 
 
 Fran:e 7 days, 284 pounds ; 28 days, 497 pounds 
 
 (Cahier Gullian) 
 
 Russia 7 days, 255 pounds (Russian Public Works 
 
 Belgium 7 days, 255 pounds ; 28 days, 497 pounds 
 
 (Genie Beige) 
 
 England 7 days, 350 pounds (Faija) 
 
 Germany No neat requirements 
 
 American Soc. of C.E.. . 250 to 550 pounds, 350 to 700 pounds. 
 
 (tests recommended) 
 
 Sand Tests. 
 3 Sand to i Cement. 
 
 France 28 days, 215 pounds 
 
 Germany ,, ,, 227 ,, 
 
 Russia ,, ,, 142 ,, 
 
 Belgium , ,, 215 ,, 
 
 England None 
 
 American Society of C.E. recommends 7 days, 80 to 125 pounds,. 
 28 days, 120 to 200 pounds 
 
 Checking. After three years' discussion, continued Mr. 
 Lesley, the boiling test was finally reported against at the 
 Session 1893, by the German Cement Association, in con- 
 junction with the Prussian Minister (Gray's Paper, World's- 
 Fair Congress, August, 1893). Nor is it contained in the 
 most notable French specifications. Faija's Paper, at the 
 World's Fair Congress, August 2nd, 1893, recommended 
 the steaming test only, and immersion at 116 deg. Fah. 
 The American Society of C. E. requires no boiling test. 
 
 Mr. Grants Results. The following conclusions weie 
 arrived at by the late Mr. John Grant, M. Inst. C.E., in 
 the course of his very extensive and exhaustive experiments, 
 on the strength of cements : 
 
 (i) The earlier experiments, made six years ago (1859),. 
 
9 2 
 
 .gave, at the end of a month, a breaking weight varying from 
 75 Ih. to 719 Ib. on an area of z\ square inches. 
 
 (2) A minimum test of, at first, 400 Ib., and afterwards of 
 500 Ib., was specified. 
 
 (3) During six years the average strength of 1,369,210 
 bushels used in the Southern Main Drainage Works has 
 been 606 '8 Ib., being 52 per cent, above the standard first 
 specified, and 21 per cent, above that subsequently adopted. 
 
 (4) The average weight per bushel has been ii4*i5lb., 
 being 4-15 per cent, above the specified standard. 
 
 (5) Portland cement has been proved to be peculiarly 
 suitable for hydraulic works, and may be procured in any 
 quantity and of the highest quality. 
 
 (6) Portland cement, if it be preserved from moisture, 
 does not, like Roman cement, lose its strength by being 
 kept in casks or sacks, but rather improves by age a great 
 
 .advantage in the case of cement which has to be exported. 
 
 (7) The longer it is in setting the more its strength 
 increases. 
 
 (8) Neat cement is stronger than any admixture of it with 
 .sand. 
 
 (9) Cement mixed with an equal quantity of sand (as has 
 been the case throughout the Southern Main Drainage 
 Works) may be said to be, at the end of a year, approxi- 
 mately three fourths of the strength of neat cement. 
 
 (10) Mixed with two parts of sand it is half the strength 
 of neat cement. 
 
 (n) With three parts of sand the strength is a third of 
 neat cement. 
 
 (12) With four parts of sand the strength is a fourth of 
 11 eat cement. 
 
 (13) With five parts of sand the strength is about a sixth 
 a t cement. 
 
93 
 
 (14) The cleaner and sharper the sand, the greater the 
 strength. 
 
 (15) Very strong Portland cement is heavy, of a -bluish- 
 grey colour, and sets slowly. Quick-setting cement has 
 generally too large a proportion of clay in its composition, 
 is brownish in colour, and turns out weak, if not useless. 
 
 (16) The stiffer the cement mortar, that is, the less the 
 amount of water used in working it up, the better. 
 
 (17) It is of the greatest importance that the bricks or 
 stone with which Portland cement is used should be 
 thoroughly soaked with water. If under water in a quies- 
 cent state, the cement will be stronger than cut of water. 
 Table XIX. shows that cement kept in water was one-third 
 stronger than that kept out. 
 
 (18) Blocks of brickwork or concrete made with Port- 
 land cement, if kept under water till required for use, would 
 be much stronger than if kept dry. 
 
 (19) Salt water is as safe for mixing with Portland cement 
 as fresh water. 
 
 (20) Bricks made with neat Portland cement are as 
 strong at from six months to nine months as the best quality 
 of Staffordshire blue bricks, or similar blocks of Bramley 
 Fall stone, or Yorkshire landings. 
 
 (21) Bricks made of 4 parts or 5 parts of sand to i part 
 of Portland cement will bear a pressure equal to the best 
 picked stocks. 
 
 (22) Portland cement concrete, made in the proportions 
 of i of cement to 8 of ballast, in some cases, and of i to 
 6 in others, has been extensively used for the foundations 
 of the river wall, piers of reservoirs, and foundations 
 generally, at Crossness and Deptford, with the most perfect 
 success, and it is believed that it might be much more 
 extensively used as a substitute for brickwork or masonry- 
 
94 
 
 wherever skilled labour, stone, or bricks are scarce, and 
 foundations are wanted at the least expenditure of time or 
 money. 
 
 (23) Wherever concrete is used under water, care must 
 T^e taken that the water is still; as otherwise a current, 
 whether natural or caused by pumping, will carry away the 
 cement and leave only the clean ballast. 
 
 (24) Roman cement, though about two-thirds the cost ot 
 Portland, is only about one-third its strength, and is there - 
 fore double the cost, measured by strength. 
 
 (25) Roman cement is very ill-adapted for mixing with 
 
 In conclusion, Mr. Grant, whilst recommending Portland 
 -cement as the best article of its kind that can be used by 
 the engineer or the architect, warned anyone against its use 
 who is not prepared to take the trouble or incur the trifling 
 -expense of testing it, as if manufactured with improper 
 proportions of its constituents chalk and clay or im- 
 properly burnt, it may do more mischief than the poorest 
 lime. 
 
 Mechanical Tests of Portland Cement. Various machines 
 have been introduced for the purpose of testing the tensile 
 strength of cements by breaking briquettes formed under 
 definite conditions, at stated intervals. The two most 
 commonly employed, viz., that of the late Mr. Henry Faija, 
 constructed by Messrs. Bowes, Scott, and Western, Limited, 
 .and that of Messrs. George Salter and Co., of West Brom- 
 wich, are described below. The following instructions for 
 testing were drawn up by the late Mr. Faija : 
 
 "Sampling. When it is required to take a sample of 
 cement for testing, it is desirable, in order to secure a fair 
 average sample, to take a small quantity from two or three 
 out of every h mdred barrels or sacks, or the equivalent in 
 
95 
 
 bulk, and mix them ail well together before taking the 
 quantity required for testing, the samples being taken well 
 from the centre of the sacks, barrels, or bulk, and not from 
 the surface, as that portion may have been accidentally 
 damaged. 
 
 " Gauging and Manipulation. To obtain the best results, 
 whether in the testing-room or in actual work, the minimum 
 of water should in all cases be used when gauging cement. 
 In the testing-room, small experimental pats should be 
 made with a weighed quantity of cement and a measured 
 quantity of water, in order to determine the exact amount 
 of water required to properly gauge the particular sample 
 under consideration. Having arrived at this knowledge, a 
 sufficient quantity of cement for filling a nest of moulds 
 (five moulds require 30 oz.) should be put into the gauger, 
 and the proper quantity of water added thereto. The 
 handle of the gauger must then be turned until the cement 
 has attained the proper consistency ; it should then be turned 
 out of the gauger on to the gauging slate, knocked up with 
 a trowel into a convenient form, and filled into the moulds, 
 being lightly rammed and gently shaken in order to remove 
 all air bubbles. The briquettes should then be smoothed 
 off with a trowel and placed on one side. The whole 
 operation, from the time of putting the water to the cement 
 to placing the briquettes on one side, should not exceed 
 five or six minutes. The briquettes should be removed 
 from the moulds at the expiration of twenty-four hours 
 from gauging, and placed in water, where they should 
 remain until due for testing. It is customary to determine 
 the tensile strength at the different dates by the average 
 of five briquettes at each date. 
 
 " Three pats should be made on pieces of glass or other 
 non-porous substance, and their behaviour watched under 
 
9 6 
 
 the following conditions : Pat No. i may be left in the air> 
 and pat No. 2 should be put in water as soon as it is set. 
 Pat No. 3 should be treated in the apparatus for deter- 
 mining the soundness of cement. The apparatus consists 
 of a covered vessel, in which water is maintained at an 
 even temperature of nodeg. The space above the water 
 is therefore filled with the vapour rising therefrom, and is- 
 at a temperature of about 100 deg. Immediately the pat is- 
 gauged it should be placed in the upper part of the vessel,. 
 in which racks are provided for the purpose, and in five or 
 six hours it may be placed in the warm water and left 
 therein for nineteen or twenty hours. If, at the end of that 
 period, the pat is still fast to the glass and shows no signs 
 of blowing, the cement may be considered perfectly sound ; 
 should, however, any signs of blowing appear, the cement 
 should be laid out in a thin layer for a day or two, and a 
 second pat made and treated in the same manner, as the 
 blowing tendency may only be due to the extreme newness 
 of the cement. If pat No. 3 shows the cement to be 
 unsound, pats No. i and 2 will eventually prove it, but it 
 may be weeks or even months before they develop the 
 characteristics. If pat No. 2 blows, it may be because it 
 was put into the water before it was set. 
 
 " SPECIFICATION. 
 
 " Fineness. To be such that the current will all pass 
 through a sieve having 625 holes (25^) to the square inch, 
 and have only 10 per cent, residue when sifted through a 
 sieve having 2500 holes (5o 2 ) to the square inch. 
 
 ^Expansion or Contraction. That a pat made and sub- 
 mitted to moist heat and warm water at the temperatures 
 and in the apparatus already described, shall show no signs 
 of blowing in twenty-four hours. 
 
97 
 
 " Tensile Strength. Briquettes which have been gauged, 
 treated, and tested in the prescribed manner, to carry an 
 -average tensile* strain, without fracture, of at least 175 Ib. 
 at the expiration of three days from gauging, and those 
 tested at the expiration of seven days from gauging, to show 
 an increase of at least 50 per cent, over the strength of 
 those at three days, but to carry a minimum of 350 Ib. per 
 square inch. 
 
 "In applying the strain to a briquette when testing for 
 tensile strength, it is important that the strain should be 
 applied evenly and always at the same rate. A difference 
 of 25 per cent, may be obtained by applying the strain very 
 quickly or very slowly." 
 
 In the " Proceedings," Institution of Civil Engineers, 
 vol. Ixxv., page 225, Mr. Faija gives the following : 
 
 "Summary of Results of Experiments to Determine the Difference Obtained 
 by Applying the Weight to the Briquette, when Testing for Tensile 
 Strength at Different Speeds. 
 
 No. of briquette. 
 
 129 .. . 
 
 129 . . . 
 
 145 . . . 
 
 145 . . . 
 
 90 .. . 
 
 90 .. . 
 
 40 .. . 
 
 40 .. . 
 
 " From the foregoing results it will be seen that the in- 
 crease per cent, due to increased speed of applying the 
 -strain is as follows : 
 
 " Taking the lowest speed of 100 Ib. in 120 seconds 
 as a starting-point, by applying the strain at the rate of 
 
 H 
 
 Speed. 
 
 Average result 
 
 Pounds. Seconds. 
 
 Pounds. 
 
 100 in i 
 
 .. .. 560-75 
 
 100 in 15 . . 
 
 .. .. 506-43 
 
 100 in 15 
 
 .. .. 452-20 
 
 100 in 30 
 
 .. .. 430-96 
 
 100 in 30 
 
 .. .. 4 I 7'27 
 
 100 in 60 
 
 .. .. 403-05 
 
 100 in 60 
 
 .. .. 416-75 
 
 100 in 120 . . 
 
 . . . . 400 87 
 
9 8 
 
 100 Ib. in 60 seconds; the increase is 3-960 per cent. 
 
 100 Ib. in 30 7-488 ,, 
 
 100 Ib. in 15 12-416 ,, 
 
 100 Ib. in i 23-142 
 
 " The standard of speed now adopted is 100 Ib. per 
 15 seconds." 
 
 The following is an illustration of Faija's cement-testing 
 machine. 
 
 The ordinary-sized machine, adapted to test briquettes of 
 i square inch section, will test from i Ib. up to 1000 Ib.; it 
 stands i4in. high, is i4in. long by 3111. wide, and weighs 
 under 30 Ib. 
 
 The knife edges and wearing parts are of phosphor 
 bronze, and special, gearing has been arranged so that the 
 strain may not be put on the briquette at too great a 
 
99 
 
 speed: 100 Ib. per 15 seconds is now considered the 
 standard. The clips of the machine are made to suit the 
 form of briquette adopted by the Metropolitan Board of 
 Works. The clips can be made to suit other forms of 
 briquette as required. 
 
 The. makers' instructions for using this machine are as- 
 follows : On receiving the machine, clean off all old oil 
 and re-lubricate ; attach the balance weight W to the short 
 end of the lever. 
 
 71? Use the Machine. See that the quadrant A is in the 
 position shown in sketch, so that the chain B to the dial C 
 is slack, and the lever D free and balanced. 
 
 Turn the wheel E from right to left, until the lower clip 
 F can be raised into contact with the upper clip G. 
 
 Insert the briquette to be tested in the clips, taking care 
 that it is put in true and evenly, so that the pull en it and 
 the clips is true and vertical ; then turn the wheel E from 
 left to right, which will bring down the lower clip F y and 
 secure the* briquette firmly in the clips. (It is generally 
 advisable to put such a strain on the briquette by turning 
 wheel E, that about 100 Ib. is indicated on the dial.) When 
 in this position there should be about half an inch between 
 the under side of knife edge H and the buffer or recpil 
 spring I. 
 
 Having seen that the pinion K is in gear with the wheel 
 L, turn the handle M until the briquette breaks. The 
 loose pointer will show on the dial the strain in pounds at 
 which the briquette broke. 
 
 To Return to Ztro. Throw the pinion K out of gear 
 with the wheel L by removing the pin and pushing it to the 
 left ; turn the wheel L from left to right until the quadrant 
 A has returned to its normal position, with the chain B 
 slack ; put the loose pointer back to zero ; release the lower 
 
 H 2 
 
100 
 
ICT 
 
 clip F by turning wheel E from right to left ; remove the 
 broken briquette, and insert the next that is to be broken. 
 
 Messrs. Geo* Salter and Co., of West Bromwich,' are the 
 makers of a Compound Lever Cement Testing Machine, 
 of which a woodcut is given on the page opposite. 
 
 This apparatus consists of a cast iron column, carrying rt 
 pair of compound levers, having a combined leverage of. 
 50 to i. The levers are fitted with tempered steel knife 
 edges, which rest on polished concave bearings, also of 
 tempered steel, thus obtaining a very sensitive balance. 
 A sliding balance-weight, for the purpose of setting the 
 levers in equilibrium, is fitted to the upper lever. The 
 upper clamp to receive the cement briquette is suspended 
 from a knife edge on the lower lever ; the lower clamp is 
 attached to the base of the corumn, and is adjustable by 
 means of a screw and a small hand-wheel. The supply of 
 shot to the bucket is automatically cut off at the moment 
 the briquette breaks. To use the machine, set the levers 
 " floating " by means of the sliding balance-weight W, then 
 place the briquette in the clamps, hang the bucket on the 
 bridle B, and screw down the small hand-wheel until the 
 end of the lever from which the shot bucket is suspended 
 is as high as it can conveniently be ; press down the handle 
 H until the shot flows at the desired rate. Immediately 
 the briquette breaks, the bucket into which the shot has 
 been poured falls upon the lever foot F, the sliding shutter 
 S is released, falls, and cuts off the supply of shot. The 
 bucket and shot are then weighed on a Salter's Spring 
 Balance, fitted with a special dial, on which the breaking 
 strain is indicated without any calculation. Should the shot 
 become blocked, and cease to flow, a slight tap will re-start 
 it. A mixture of shot, No. 6 and No. 10, may be used. The 
 maximum load on the machine should not exceed 1000 lb.- 
 
IO2 
 
 Little is really known yet of the action of perco- 
 lation of water (fresh, rain, or salt) or paraffin, &c., 
 through various strengths of concrete and under various 
 heads. 
 
 To endeavour to find out the most economical concretes 
 for this, Messrs. Hilton and Co. have devised a neat appa- 
 ratus for the purpose, and easy for anyone to use. It is 
 made solely by Calvert, Harris, and Co , of 54, Cannon 
 Street, London. 
 
 It consists of a small hand pump over a cistern, which 
 pumps up an accumulator, the table of which can be 
 weighted to any head desired, and which is read at a glance 
 by the special gauge marked off into feet head and pounds 
 per square inch. This is connected to two gun-metal boxes, 
 heavily ribbed to prevent bulging and leaking when under 
 pressure. The interior of the boxes are also slightly ribbed, 
 to prevent all chance of leakage next the metal. Each box 
 measures a foot cube exactly, and either can be shut off the 
 pressure pipe and be entirely disconnected or not at will. 
 Foot cube blocks can be reported on, or any less thickness, 
 or the top and bottom of the case can be bolted together 
 over any sample paving slab, and results equally well noted 
 The percolation runs through into the bottom half of box, 
 and is measured in special closed glass containers marked 
 off in cubic centimetres. 
 
 The inventors of this machine will shortly have some 
 "useful information to impart on percolation, as they are 
 continuing their experiments in this direction. 
 
 Two boxes are supplied to enable another sample to be 
 got ready while one is under test, or a different quality 
 sample can be tested at same time in the second box, and 
 so time saved. More boxes can be coupled up to the same 
 accumulator if desired, to still further save time. Under 
 
the best circumstances making and recording these tests is 
 necessarily a slow and tedious process, and requires great 
 care. 
 
104 
 
 CHAPTER X. 
 
 THE ADULTERATION OF PORTLAND CEMENT. 
 
 THE great importance of cement being thoroughly reliable 
 has naturally led to considerable discussion as to the necessity 
 of this substance being supplied in a pure and sound condi- 
 tion, and the practice which has arisen of adding to good 
 cement various ingredients with a view to its so-called 
 improvement has been sharply called to account. 
 
 So injurious to the interests of the manufacturers of good 
 cement has this sophistication been, that in 1894 the cement 
 trade held a meeting at the London Chamber of Commerce* 
 " for the purpose of establishing an association of English 
 cement manufacturers," and "of dealing with, and, if possible, 
 of putting a stop to, the unprincipled and disreputable prac- 
 tice " of " adulterating cement by the mixture of Kentish 
 rag-stone, other stone, furnace or oven ashes, disused or 
 exhausted firebricks, and other inert material." It was attended 
 by representatives of twenty-nine manufacturers of Portland 
 cement. There were considerable differences of opinion on the 
 various points, some approving and others objecting to the 
 proposed association, whilst others preferred to leave the 
 matter in the hands of the London Chamber of Commerce. 
 One gentleman denied that the addition of Kentish rag- 
 stone was an adulteration, and contended that it improved 
 the cement in point of colour and tensile strength, the sand 
 tests were higher, and in every respect it was a better 
 
 * Journal of the Society of Chemical Industry, December 3ist, . 
 1894, page 1236. 
 
article. Another manufacturer said that English cement 
 manufacturers were making a better article than they did. 
 some years ago,but admitted that they had been distanced 
 by German skill. He thought they " ought not to be 
 debarred from imitating that skill, from taking advantage of 
 scientific research, and from making all the progress they 
 could ; and therefore held that the matter should be sifted 
 by a qualified body in a scientific manner, and then fully 
 reported upon." 
 
 " In this connection it is instructive to note," continues 
 the Journal, " what the Germans have done in the same 
 direction. According to the Chamber of Commerce Journal, 
 early in the year 1880, when adulteration threatened to 
 become very prevalent, the Union of German Cement 
 Manufacturers found itself compelled to set its face in the 
 strongest possible manner against the admixture of foreign 
 materials with cement, in order to protect both their good 
 name as well as the building trade, from disadvantage and 
 danger. On the other hand, the practice was defended as 
 producing a cheaper cement and yet sufficiently strong for 
 that trade. The Union, however, insisted that the question 
 of price must be left to the consumer, who, if he wished,, 
 could make the mixtures himself. If manufacturer or dealer 
 were permitted to do this, the door would be open wide to- 
 fraud, and the confidence of the public be shattered, for 
 the consumer would not be in a position to know or test 
 either the degree of adulteration or the characteristics of" 
 the added materials. 
 
 " With regard to the alleged improvement, it was certainly 
 known that by the addition of ultramarine, precipitated 
 silicic acid, preparations of potters' clay, &c., the strength of 
 Portland cement could be increased, but it was still 
 doubtful whether the other valuable qualities of Portland 
 
io6 
 
 cement do not suffer, and, moreover, these admixtures are 
 excluded, if only by reason of their high price. 
 
 " With the view of rooting out, as far as possible, this 
 system of adulteration, the Union drew up certain informa- 
 tion as to what may be legitimately added to cement and 
 what may not. The following were the conclusions arrived 
 at : 
 
 "Admixture of Gypsum. The addition of unburnt 
 gypsum, made at the time of grinding, has for its advanta- 
 geous object the retarding of the setting of such Portland 
 cements as are by their nature quick-setting, and thereby 
 rendering them more fit for use. The action of the gypsum 
 is probably owing to the fact that when the cement is mixed 
 with water into mortar, the gypsum is first dissolved, and 
 then precipitated in extremely fine particles on the grains of 
 the cement, in exactly the same way as it is recommended 
 for lime by F. Schott (see Dingier* s Journal, 205, 52, and 
 209, 30). If the particles of cement are enveloped in 
 'the thinnest possible film of gypsum, the chemical incorpo- 
 ration of the water is notably delayed. 
 
 " In consequence of this, the cement becomes slower in 
 setting, and by this means (possibly also by the simultaneous 
 occurrence of some chemical action) cement mixed with a 
 little gypsum gives a higher degree of strength than the 
 unmixed quick-setting cement. Inasmuch as the addition 
 of from J to 2 per cent, of gypsum is sufficient for the 
 attainment of this object, and as, moreover, the addition of 
 a larger percentage would cause the cement to 'blow,' 
 there can be no possible ground for asserting that the 
 admixture of gypsum is made with a fraudulent intention ; 
 nay, more, such an addition, within the prescribed limits, 
 must be regarded as an improvement. In consequence of 
 this, the united declaration of the German Portland cement 
 
icy 
 
 manufacturers specially permits additions, rot exceeding 
 2 per cent, if made with the object of regulating the time 
 of setting. 
 
 "Fraudulent Admixtures. Of the materials which, may 
 from time to time be mixed with Portland cement with the 
 intention of adding to the profits, and without open acknow- 
 ledgment of such mixture, it is evident that only those can 
 be employed which are cheap, and which resemble Portland 
 cement both in weight and colour. Chief amongst such 
 materials are slag principally blast furnace slag grey 
 limestone, slate, hydraulic lime, black or grey limestone, 
 and trass. The most popular adulterant is slag, because, 
 when pulverised, it is very similar in appearance to Portland 
 cement, and is therefore less easily detected, even when 
 added in large proportions. 
 
 " All the foregoing admixtures make the cement inferior, 
 not only in respect of strength, but also in respect of those 
 other valuable qualities (or characteristics) of Portland 
 cement, such as resistance to frost, durability, cohesiveness, 
 &c., as has been proved by the most exhaustive researches. '' 
 
 As a result of the agitation thus started, the London 
 Chamber of Commerce, on the 3oth November, 1894, 
 requested Mr. W. Harry Stanger, M. Inst. C.E., and Mr. Bertram 
 Blount, F.I.C., to undertake the investigation of the effect on 
 Portland cement of the admixture with it of various foreign 
 substances, especially of Kentish rag-stone, and on the 1510 
 May, 1896, those gentlemen presented the final results of 
 their investigations to the Chamber. 
 
 As a result of their labours, the Special Committee of the 
 Cement Trade Section of the London Chamber of Com- 
 merce carried a resolution to the following effect : 
 
 " That Portland cement be defined as a mixture of two 
 or more suitable materials, intimately and artificially mixed 
 
io8 
 
 in the requisite proportions, and afterwards properly calcined 
 and ground, to which nothing has been added during or 
 after calcination, excepting that an addition not exceeding 
 2 per cent, of gypsum is permissible for the purpose of 
 regulating the setting." 
 
 So important, state Messrs. Stanger and Blount, did it 
 appear to the members of the Section to secure Portland' 
 cement from adulteration, that they drew up the following 
 declaration, which all manufacturers of Portland cement in 
 Great Britain and Ireland were to be invited to sign : 
 
 " We, the undersigned, hereby agree to conform to and! 
 carry out the rule of the Cement Trade Section of the- 
 London Chamber of Commerce, as set forth in a report 
 made by the Section, and adopted at a meeting held on- 
 Monday, the loth May, 1897 : 
 
 " That if any material whatever, excepting an amount not 
 exceeding 2 per cent, of gypsum for the purpose of regu- 
 lating the setting, be added to Portland cement clinker,, 
 during or after calcination, the article so produced shall not 
 be sold as Portland cement, but under some other distinc- 
 tive name. 
 
 " And we further agree that if at any time any of the 
 parties to this agreement shall, by resolution of a majority 
 of those present at a meeting of such parties duly and 
 properly convened in accordance with the practice of the 
 London Chamber of Commerce, such resolution having; 
 been duly and properly confirmed by a majority of those- 
 present at a subsequent meeting, called at not less than 
 fourteen days' notice, be found to have failed to conform to 
 and carry out the said rule, then in such case his or their 
 name or names shall be struck off the list, and notice of the 
 same made public in such manner as shall be resolved." 
 
 This placed the matter beyond controversy, but in order 
 
that my readers may understand the reasons for this decisive 
 step, I am permitted to append the following substance of 
 Messrs. Stanger and Blount's report to the Chamber of 
 Commerce, a precis of which was published in the form of a 
 paper read before the Society of Chemical Industry on 
 November ist, 1897, above referred to, as follows : 
 
 " Outline of the Investigation. 
 
 " On the question arising as to the effect of an addition 
 of ragstone to Portland cement, it would appear fairly 
 obvious, on chemical grounds, that the ragstone would be 
 an inert substance, and could be regarded merely as a 
 diluent. Kentish ragstone is a natural form of carbonate of 
 lime mixed with siliceous matter. It varies somewhat in 
 composition, as may be seen from the analyses in Table I., 
 which are calculated on the samples free from moisture. 
 I3ut although thus varying, its composition is always such 
 that when the stone is powdered and mixed with water it is 
 In no way cemefltitious, and shows no tendency to set 
 
 * Nevertheless it was stated, by those that had made the 
 experiment, that this inert, silicious limestone could be 
 added to Portland cement in considerable quantity (10-20 
 per cent.) with positive advantage to the cement. It was 
 credibly asserted that a cement thus diluted was actually 
 stronger than one in its normal unmixed state. 
 
 " Rejecting both a priori views and the statements of other 
 <xperimenters, w r e made our own tests. We may say here 
 that in all cases our experiments have been made on cement 
 and ragstone, ground together as in actual manufacture ; 
 separate grinding of the materials and subsequent admixture 
 not only departs from the method practised by users of 
 ragstone, but is likely to yield misleading results, because 
 the intimacy of admixture is insufficiently great. Mixtures 
 
no 
 
 DIAGRAM OF TENSILE TESTS. SERIES "A.' 
 
 ea ? 
 
 _ ssfisfiS^HIg 
 
 ^^^ 1i 
 
 /*#> cM1 ro.- -'"-'" SsJj J 
 
 iro 
 
 7 OATS. ?6 DAYS. 
 
Ill 
 
 containing 10, 20, and 50 per cent, of ragstone were 
 prepared in this manner, and tested at seven and twenty- 
 eight days, and aj: six months, neat and with sand, in tension 
 and compression, and compared with corresponding tests 
 made on the unmixed cement. The diagram on page 108 
 is typical of certain of the results obtained. 
 
 "The composition of these samples is stated in Table II. 
 
 " Many similar tests proved beyond doubt that some 
 cements not only do not show a decrease in strength when 
 mixed with 10 or 20 per cent, of ragstone, but are actually 
 stronger when thus mixed. This is true whether the cement 
 and mixtures are tested neat or with sand, in tension or in 
 compression. Thus the upshot is in direct contradiction to 
 what would certainly be expected by anyone endeavouring, 
 to solve the question on first principles. The result consti- 
 tutes a useful example of the danger of making even what 
 appear to be most reasonable assumptions, when it is open 
 to one to dispense with assumptions altogether, and to sub- 
 stitute sound experimental data. 
 
 u After close study, the reasons for this apparent anomaly 
 were discerned. It became clear that when ragstone con- 
 taining, as it commonly does, a small quantity of moisture, 
 is mixed with cement clinker and passed, together with that,, 
 through the stone-breaker and millstones, or other grinding 
 machinery, it is brought into the closest possible contact 
 with the cement, and slakes any overlimed portions which 
 may be present. Thus, a cement containing ragstone as it 
 comes from the mills is more slaked than one free from 
 ragstone. Now it is known that cement which is not 
 perfectly sound is improved by limited slaking ; it is on 
 account of this that air-slaking or aeration is insisted on by 
 many large users (notably by the Admiralty) in order to 
 hydrate any uncombined lime or unstable lime compounds,. 
 
112 
 
 
 g 
 
 
 
 00 
 
 !_!, 
 
 Q O VO ^r M G'O ro ro roC^iO 
 
 o 
 
 in N n 
 
 
 
 - 
 
 ** o in o >~* "^ o o in o o 
 
 o 
 
 ^- o o 
 
 0< 
 
 
 PH M 
 
 o 
 
 ^ 00 
 
 
 =. 
 
 OOThONwCO Tt-COOO 
 
 8 
 
 (N ^ ro 
 
 1 
 
 _ 
 
 '-'OinoiHinoo ro M o 
 
 
 
 in o o 
 
 <N 
 
 
 __i___r 
 
 2 
 
 
 
 (xi 
 
 jj ? o j o j o ! 
 
 .8 
 
 0\ ro r^ 
 VO <N N 
 
 1 
 
 
 PL, N ^ ^ 
 
 
 
 2 
 
 M in o 
 
 N 
 
 Q 
 
 /Rooooooo o ^ o 
 
 8 
 
 o in M 
 ro c^ in 
 
 1 
 
 
 ' H Oino>-iinoo ^MO 
 
 o 
 
 in r^ o 
 
 M 
 
 
 rv M ^ ^ 
 
 o 
 
 
 
 
 +1 
 
 
 
 O 
 
 ^ 
 
 F] ^O ro ^t~ N *^t" O O ^ *O 
 
 
 
 ro 
 
 
 ^J 
 
 ^TfiO'^-C4COrO'-H OO^O 
 
 o 
 
 i ^ 1 
 
 t^ 
 
 5 
 
 ^^' ( ^- ro 
 
 8 
 
 00 
 
 M 
 
 
 
 
 o 
 
 
 
 
 
 H 
 
 o 
 
 
 
 
 
 
 
 
 
 
 rt 
 
 
 
 
 
 s 
 
 
 
 
 
 
 * 
 
 
 
 
 ^ i 
 
 
 
 C/3 
 
 
 
 
 " 
 
 
 O 
 
 
 V-< r* 
 
 
 
 
 
 
 O " f \ w 
 
 * 
 
 
 d 
 
 
 ^ ^r 
 
 
 
 * ' -^ 
 
 
 1 g 
 
 * 
 
 
 
 . 
 
 o : J : 
 
 
 
 ? i ^ ! & 
 
 
 C rt <u 
 
 
 
 . | ; . ; . 1 | ; | 
 
 . 
 
 O * '-; en 
 
 
 
 : 1 : : : 7 1 ^ : 5 
 
 ; 
 
 C g 
 
 .2 ^ X =l 
 
 rt 
 
 
 :^~< : o'tsfffi| 
 
 ; 
 
 ? g ^ 
 
 l| 3 
 
 r-=1 3 C W 
 
 tao 
 o 
 
 
 *~ i ^-i. ^^ ^ Q^ ^ r^H o t^ *-* 
 
 03 
 
 ^ w ^ rt 
 
 o 
 
 
 ^=5<D^ o>-^<5 rt"rt'S (2 
 
 O 
 
 ^ e ^ 1-1-1 
 
 <U G C ^3 
 
 & 
 
 
 "J ^s 'I -8 T a -| So^^ 
 
 H 
 
 cS S 
 
 OS-Q PH ^ 
 
 o rt j 3 
 3 b 
 
 in 
 
 
 
 
 rt o rt 
 > U O 
 
 # - 
 
 
TABLE II. 
 
 *. - 
 
 Un- 
 mixed 
 cement. 
 
 IO 
 
 per cent 
 admix- 
 ture. 
 
 20 
 
 per cent 
 admix- 
 ture. 
 
 50 
 per cent, 
 admix- 
 ture. 
 
 ica (SiO 2 ) 
 
 Per cent 
 20-32 
 
 Per cent 
 18-36 
 
 Per cent 
 I5-54 
 
 Percent. 
 10-98 
 
 Insoluble residue (sand & clay) 
 
 0-70 
 
 2'34 
 
 4-28 
 
 7-54 
 
 Alumina (A1 2 O 3 ) 
 
 8-36 
 
 7-16 
 
 6'22 
 
 4-72 
 
 Ferric oxide (Fe 2 O 3 ) . . 
 
 4'34 
 
 4-18 
 
 3'80 
 
 2-80 
 
 Lime (CaO) 
 
 59-4^* 
 
 58-72* 
 
 56-62* 
 
 53-io* 
 
 M agnesia (MgO) 
 
 1-04 
 
 i* 02 
 
 I -O2 
 
 0-83 
 
 ulphuric anhydride (SO 3 ) . . 
 
 i'37t 
 
 I'26f 
 
 I-Ilf 
 
 o*75t 
 
 Matter volatile at a red heat : 
 
 
 
 
 
 Carbonic anhydride (CO 2 ) 
 
 1-77 
 
 4-90 
 
 9-09 
 
 17-42 
 
 Water (OH 2 ) 
 
 i-Si 
 
 1-64 
 
 1-81 
 
 i-34 
 
 kalis: Potash {KjO), soda 
 (N^O), and loss .. .. 
 
 0-83 
 
 0-42' 
 
 0-51 
 
 0-52 
 
 
 loo-oo 
 
 100-00 
 
 lOO'OO 
 
 100-00 
 
 ''Value after deduction of that 
 necessary for the forma- 
 tion of calcium sulphate . . 
 
 53-5 
 
 57-84 
 
 55-84 
 
 52-58 
 
 Calculated as calcium sul- 
 phate . 
 
 2 . TO 
 
 2 ' 14. 
 
 I -80 
 
 I ' 27 
 
 
 * JO 
 
 * *T I 
 
 * uy 
 
 * */ 
 
 Specific gravity 
 
 3-075 
 
 3'06O 
 
 2'975 
 
 2-920 
 
 FINENESS. 
 esiduo on 76 x 76 sieve . . 
 
 i 
 
 9 
 
 8J 
 
 8 
 
 50 x 50 
 
 4i 
 
 4 
 
 4 
 
 4 
 
 25 x 25 ,, 
 
 i 
 
 * 
 
 i 
 
 i 
 
DIAGRAM OF COMPRE8SIVE TESTS. SERIES "A." 
 
 CUBES 3'X3'X3' 
 
 28 
 
 DAYS. 
 
IJ 5 
 
 and prevent their subsequent hydration and destructive 
 expansion when the cement as a whole is gauged with water 
 for preparing concrete or mortar. A moist ragstone, ground 
 
 NEAT TESTS. 
 
 Breaking Weight in Pounds of Briquettes having lin. x lin. Section. 
 
 Unmixed Cement. 
 
 10 per cent. Admixture. 
 
 520 
 490 
 
 460 
 
 550 
 540 
 530 
 
 480 
 470 
 410 
 
 5io 
 500 
 480 
 
 Average 490 Ib. 
 
 Average 540 Ib. 
 
 Average 453^ Ib. 
 
 Average 496-3 Ib. 
 
 
 20 per cent. Xdmixture. 
 
 50 per cent. Admixture. 
 
 7 days. 21 days. 
 
 7 days. 
 
 21 days. 
 
 410 
 390 
 
 380 
 
 410 
 410 
 390 
 
 240 
 
 220 
 220 
 
 300 
 290 
 250 
 
 Average 393^ Ib. 
 
 Average 403^ Ib. 
 
 *\verage 226^ Ib. 
 
 Average 280 Ib. 
 
 with such a cement, performs this function of limited hydra- 
 tion, and thus subjects the cement to what is, in effect, an 
 artificial and rapid aeration. Cements which are not per- 
 fectly sound are improved by this procedure. But when a 
 
 I 2 
 
n6 
 
 cement is rendered perfectly sound by aeration, this 
 improvement disappears, as is evident from the diagrams 
 given on pages 118 and 119. 
 
 SAND TESTS 
 Breaking Weight in Pounds of Briquettes having lin. x lin. Section. 
 
 Unmixed Cement. 
 
 10 per cent, Admixture. 
 
 7 days. 
 
 21 days. 
 
 7 days. 
 
 21 days. 
 
 300 
 
 330 
 
 250 
 
 330 " 
 
 260 
 
 320 
 
 240 
 
 290 
 
 250 
 
 280 
 
 
 
 
 
 Average 270 Ib. 
 
 Average 310 Ib. 
 
 Average 245 Ib. 
 
 Average 310 Ib. 
 
 20 per cent. Admixture. 
 
 50 per cent. Admixture. 
 
 7 days. 
 
 21 days. 
 
 7 days. 
 
 21 days 
 
 240 
 
 300 
 
 130 
 
 200 
 
 230 
 
 270 
 
 100 
 
 1 60 
 
 Average 235 Ib. 
 
 Average 285 Ib. 
 
 Average 
 
 Average 180 Ib. 
 
 "These figures show that in an aerated, and therefore 
 perfectly sound cement, the ragstone acts precisely as would 
 be expected, viz., as a mere diluent. The effect is the more 
 striking because these very samples before aeration showed 
 
a smaller strength unmixed than when mixed with ragstone. 
 That this alteration of properties is caused merely by the 
 increased soundness of the cement, which is induced by 
 aeration, is proved by the tables on pages 115 and 116. 
 
 " In this case the cement used was not aerated by long 
 storage, and yet gave tests which were considerably lower 
 for the mixtures containing ragstone than for the unmixed 
 cement. It was not a cement made in the laboratory, and 
 therefore of better grade than can be readily manufactured 
 on a large scale, but was simply an ordinary commercial 
 cement of high quality and perfect soundness. 
 
 " A word must be said as to what is meant by the word 
 1 soundness,' which . has been used in the previous para- 
 graphs. A cement may appear perfectly sound to ordinary 
 tests, and yet after setting may develop internal stresses 
 which, though not sufficient to cause it to disintegrate or 
 show any sign of actual failure, may diminish its strength 
 appreciably. That this is not an uncommon condition is 
 clear from the fact that many cements are considerably 
 stronger after aeration than when fresh, as is evident from 
 the following table : 
 
 Breaking Weight in Pounds of Briquettes, lin. x lin. Section. 
 Unmixed Cement and Sand (3 to i). 
 
 
 7 Days. 
 
 28 Days. 
 
 
 Fresh. 
 
 Aerated. 
 
 Fresh. 
 
 Aerated. 
 
 Series "A" .. .. 
 
 i9if 
 
 2 5 6f 
 
 270 
 
 303 i 
 
 "D" .. .. 
 
 259? 
 
 2Q3J 
 
 2 7 6 
 
 383 
 
 "L" .. .. 
 
 I3i 
 
 243 
 
 190 
 
 340 
 
500 
 
 500 
 
 400 
 
 300 
 
 200 
 
 100 
 
 TENSILE TESTS AFTER AERATICN. 
 
03 
 
 o 
 
 500 
 
 400 
 
 300 
 
 200 
 
 iCC 
 
 (SAND-CEMENT BRIQUETTES 5 
 
 SERIES" F' 
 
 :0 
 
 293 
 
 143 
 
 500 
 
 400 
 
 300 
 
 200 
 
 100 
 
 7 DAYS . 
 
 fe 
 
 o 
 
 z 
 
 500 
 
 400 
 
 300 
 
 200 
 
 100 
 
 -ACE- 
 
 fsAND-CEMENT BRIQUETTES 
 
 SERIES^L" 
 
 2 s DAYS 
 500 
 
 V 
 
 212 
 
 400 
 
 300 
 
 200 
 
 1UO 
 
 *2I2. _ AGE 
 
 ENSILE TESTS AFTER AERATION. 
 
 2CDAYS 
 
I2O 
 
 " Incidentally, it may be remarked that this great 
 increase in strength is a strong argument in favour of 
 systematic and thorough aeration of all cement that is to be 
 used in heavy and important work. This has long been 
 practised by many engineers to ensure safety ; but if the 
 cement be even slightly unsound, a large increase of strength 
 will also accrue. 
 
 " Reverting to the question of the difference of strength- 
 
 UNMIXED CEMENT. 
 
 of unmixed cement and cement mixed with ragstone, it wiir 
 be observed that although experiment has established that 
 sound cement is not improved by the addition of ragstone,. 
 but is, on the contrary, deteriorated, yet the decrease of 
 strength is smaller than that which corresponds with the 
 percentage of diluent added. Thus a mixture of cement 
 with 10 per cent, of ragstone is not necessarily 10 per cent, 
 weaker, although the ragstone is merely a chemically inert 
 
121 
 
 addition The reason for this discrepancy becomes clear 
 on studying the structure of set cement. For the purpose 
 of this study, briquettes made from unmixed cement and 
 from cement mixed with 10, 20, and 50 per cent, of rag- 
 stone were chosen, and very thin sections were prepared 
 from them in the manner usual in petrological research. 
 These were examined under the microscope, and a large 
 difference in structure was at once apparent.* 
 
 IO PER CENT. ADMIXTURE. 
 
 " The microscopic section of the unmixed cement consists- 
 of white, nearly opaque, particles, interspersed with dark 
 patches fairly evenly distributed. Most of these dark 
 patches are actual gaps, but some are merely plates of 
 
 * The preparation of these photo-micrographs was kindly under- 
 taken by Mr. Herbert Jackson, of King's College, London, to whose 
 skill the beauty and accuracy of the original photographs are suffi- 
 cient witness. 
 
122 
 
 transparent crystalline substances, which, as they absorb 
 little or no light, appear as gaps in the photograph. 
 
 "A similar section of a briquette containing 10 per cent, 
 of ragstone has a much closer structure. Very few true 
 .gaps appear in it, the interstices visible in the unmixed 
 cement being filled with fine particles of ragstone. 
 
 "With 20 per cent, of ragstone a similar filling is per 
 
 2O PER CENT. ADMIXTURE. 
 
 ceptible, but there is also visible a greater proportion of 
 Jarge crystals, either transparent or opaque. 
 
 " With 50 per cent, of ragstone the filling of interstices is 
 still noticeable, but the cement is, as it were, swamped with 
 ragstone, and the section contains large inert crystalline 
 masses, which often exhibit cracks, and may be considered 
 as reducing the strength of the material. The appended 
 photo-micrographs illustrate these differences of structures. 
 
 " That the greater closeness of structure apparent in the 
 
I2 3 
 
 microscopic sections of the mixtures has a physical exist- 
 ence, was proved by making actual measurements of the 
 ;total cubic content of the cavities in the set cement and in 
 
 50 PER CENT. ADMIXTURE 
 
 'the mixtures. A maximum density occurs with the 10 per 
 cent, mixture, as is clear from the appended figures : 
 
 Volume of Cavities in a Briquette, stated in per cent, of the total 
 apparent Volume of the Briquette. 
 
 
 Series i. 
 
 Series 2. 
 
 Unmixed cement 
 
 Per cent. 
 4 -6 3 
 
 Per cent. 
 2-39. 
 
 10 per cent, mixture . . 
 
 , 3-83 
 
 i-So 
 
 20 ,, ,, 
 so 
 
 | 6-32 
 
 IO * 43 
 
 3'79 
 6-32 
 
 
 
 
TESTS FOR CONSTANCY OF VOLUME. SERIES"!."." 
 
 WET BAR SHOWN BY LINES. DRY BAR SHOWN BY LINES 
 
 ON LEFT HALF OF DIAGRAM ON RIGHT HALF OF DIAGRAM. 
 
 * INCREASE DECREASE. - 
 
 7 DAYS 
 
 T 
 
TESTS FOR CONSTANCY OF VOLUME. SERIES "11." 
 
 "WET BAR SHOWN BY THE & LINES 
 
 NEAREST IZFT HAND SIDE OF DIAGRAM 
 
 < INCREASE 
 
 DRY BAR SHOWN BY THE 8 LINES 
 NEAREST RIGHT HAND SIDE OF DIAGRAM. 
 
126 
 
 It will be seen that although the large quantity of ragstone- 
 in the 50 per cent, mixture communicates to it a loose and 
 porous structure, yet the 10 per cent, mixture perceptibly 
 exceeds the unmixed cement in closeness of texture, this 
 fact serving to confirm the microscopic observations. 
 
 " From both microscopical examination of the structure, 
 and from measurement of the cavities in briquettes of un- 
 mixed cement and of cement mixed with 10, 20, and 50 per 
 cent, of ragstone, it is evident that a mixture containing 10 
 per cent, of ragstone has a closer structure than any of the 
 others, even than unmixed cement. The ragstone, in fact, 
 during grinding, yields a quantity of fine powder which 
 serves as a ' filling ' between the crystals formed when the 
 cement sets. The absence of interstices thus caused 
 appears to compensate in some measure for the fact that the 
 ragstone is inert, and thus a cement containing 10 per cent. 
 of what is simply a diluent is not necessarily 10 per cent, 
 lower in strength than an unmixed cement. In fact, a 
 briquette, even of neat cement, may be regarded as a con- 
 crete in which the strength depends not only on the true 
 adhesive and cohesive quality of the cement, but also on 
 the exact fit of the particles of cement and those of the 
 inert matter, such as the coarse core. Any fine inert filling 
 material may improve the closeness of texture of such a 
 concrete if the average size of its particles be such as to- 
 occupy fully the spaces naturally present in set cement. 
 
 " The whole question concerning the addition of ragstone 
 may be summed up very briefly : 
 
 " i. Ragstone is not a cementitious substance, and its 
 addition to cement is an adulteration. 
 
 " 2. Perfectly sound cement is weakened by the addition 
 of ragstone. 
 
 " 3. This weakening is not fully proportional to the per- 
 
127 
 
 centage of ragstone added, because the latter acts as a fine- 
 filling material and fills up the interstices naturally presen 
 in set cement. v 
 
 " 4. Cement which is not perfectly sound, may be t^mpo- 
 rarily improved by the addition of ragstone. When the- 
 cement has become sound by aeration, this improvemen 
 disappears. 
 
 " Many minor points were examined and determined in- 
 the course of the main investigation, but the most important, 
 results are embodied in the conclusions given above. 
 
 "Additions to Cement other than Ragstone. 
 
 ' l One of these, which particularly came within our pur- 
 view in the course of our investigation for the London 
 Chamber of Commerce, is gypsum. Gypsum is used largely 
 in Germany, and to a considerable extent in this country, in, 
 quantities not exceeding 2 per cent, and usually smaller 
 than this, in ordei; to lengthen the setting time of the 
 cement. Regarding cement as a chemically finished pro- 
 duct in the state in which it comes from the kilns, needing, 
 nothing but mechanical comminution to make it saleable^, 
 the addition of any substance to the finished clinker must 
 be considered, in strictness, an adulteration. Thus gypsum 
 becomes under this definition an adulterant. Nevertheless 
 it is added for a distinct and useful purpose, and in quan- 
 tities smaller than 2 per cent, does not affect the cement 
 injuriously, as far as our own experiments indicate. The 
 average results obtained with cement containing proportions 
 of gypsum increasing from o'i to 2*0 per cent, are given in 
 the table on page 128. 
 
 " It will be seen that no harmful effect on the strength of 
 the cement can be detected. 
 
 " The effect of the gypsum on the constancy of volume of 
 
128 
 
 the cement was also determined by the measurement of test 
 bars of unmixed cement and of cement mixed with 0*1 to 
 2*0 per cent, of gypsum in a Bauschinger apparatus. The 
 results are recorded in the appended curves. On the whole, 
 a slightly greater tendency to expansion is perceptible when 
 the proportion of gypsum exceeds i per cent., but the 
 difference is not so great as to warrant the conclusion that 
 gypsum causes any dangerous amount of expansion 
 
 Breaking Weight of Briquettes, lin. x iln. section (28 Days). 
 
 
 
 Sample I. 
 
 Sample II. 
 
 Unmixed cement 
 
 Lb. 
 
 618 
 
 Lb. 
 
 800 
 
 o'l per cent, gypsum 
 
 587 
 
 693 
 
 '3 ,. 
 
 630 
 
 750 
 
 '0'5 
 
 623 
 
 777 
 
 *7 
 
 590 
 
 813 
 
 I'O ,, ,, 
 
 690 
 
 823 
 
 *'5 M 
 
 657 
 
 890 
 
 2'0 ,, ,, 
 
 613 
 
 860 
 
 " With these facts before us, it is evident that though 
 gypsum is an alien addition to Portland cement, and there- 
 fore technically an adulterant, its use can be defended on 
 the ground that it confers specific properties on the cement, 
 and does not affect its strength or soundness unfavourably. 
 The difficulty arising in the question whether cement con- 
 taining a small quantity of gypsum, added for the purpose 
 ot regulating the setting time, can be legitimately sold as 
 Portland cement, has been met by the Cement Trade 
 
I2 9 
 
 Section of the London Chamber of Commerce, by the 
 adoption of the rule which has been quoted above. This 
 rule (page icS) expressly excepts gypsum in quantities up 
 to 2 per cent, from the category of substances which, if 
 added to Portland cement, constituie an adulteration. It 
 will be noted that this convention is similar to that which 
 has been in use for some years in Germany, where it 
 appears to have worked satisfactorily. 
 
 "The last and worst adulterant with which it is our 
 purpose to deal is blast-furnace slag. As far as our experi- 
 ence goes, this most objectionable addition to Portland 
 cement is not employed on the Thames and Medway, but 
 in certain other districts it is used in large quantities for 
 the preparation of a grossly sophisticated product which is 
 'fraudulently sold as Portland cement. 
 
 . " We must not be understood as condemning true slag 
 cement made by mixing granulated blast-furnace slag with 
 slaked lime and sold under its proper title. This material 
 is a perfectly legitimate product, and has its own uses ; no 
 one can reasonably object to its utilisation if it is not 
 covertly substituted for Portland cement. But the addition 
 of blast-furnace slag to Portland cement is another matter 
 altogether. The general practice of the manufacturers who 
 seek to increase their profits by the use of slag, appears to 
 be to add to the clinker as it goes to the crushers as much 
 crude blast-furnace slag as they consider can be mixed with 
 Portland cement without risk of detection by the ordinary 
 consumer, who buys cement in quantities so small that the 
 cost of its analysis is too great for him to pay. The quantity 
 added may be as much as 30 or 40 per cent; and detection 
 is not easy, or even always possible, for an unskilled 
 observer. Apart from the fraudulent character of this 
 addition, about which no doubt can well be entertained, 
 
 K 
 
130 
 
 there arises the question of its effect on the cement. And 
 here it is necessary to make a small digression into the 
 chemistry of the subject. 
 
 " When Portland cement sets, a certain quantity of lime 
 in the hydrated state is liberated. This lime in cement 
 mortar, or concrete of fair closeness of structure remains 
 distributed throughout the mass, and is there slowly con- 
 verted into calcium carbonate. Now it is very possible that 
 this lime could be utilised more effectively if it were 
 provided with a certain quantity of silica, or an active 
 silicate with which it could unite in manner similar to that 
 of the lime of a puzzuolanic cement. Further, granulated 
 blast-furnace slag will act as a puzzuolana and unite with 
 slaked lime when the dry mixture of the two substances is 
 gauged with water. It is therefore conceivable that granu- 
 lated blast-furnace slag could be added to Portland cement 
 in such quantity that its active silicates would unite with, 
 the lime set free in the normal setting of the cement.. 
 Whether this union would be advantageous to the strength 
 of the cement is a matter for experiment. But whether it is. 
 or not, a cement thus dosed with granulated blast-furnace 
 slag could not be legitimately termed Portland cement, and 
 would have to be sold under a distinctive name. It may be 
 mentioned incidentally that ordinary slag contains a good; 
 deal of sulphur (e.g., i per cent.) in the form of calcium, 
 sulphide. This, slowly oxidising, would be likely to expandi 
 in the mass of the set cement and cause stresses, which> 
 could hardly fail to be injurious, and might be positively 
 dangerous. Thus the burden of proof that the addition of 
 granulated slag to cement is not actively harmful, rests upon 
 the advocates for its use, and even if they prove their case 
 they are confronted^by the fact that the mixture is not and. 
 cannot be Portland cement. 
 
13* 
 
 " But when the slag added to Portland cement is not 
 granulated blast-furnace slag of the best composition for 
 acting as a pmzzuolana, but is the common stony material 
 run out into trucks and allowed to cool spontaneously, 
 instead of being rapidly chilled, the objection to the addi- 
 tion of slag is even stronger. In the first place, this stony, 
 slowly-cooled slag is usually not of such a composition as to 
 allow it to act as a puzzuolana. Next, even if it were, its 
 condition i.e., annealed by slowly cooling instead of chilled 
 by quenching is most unfavourable to its puzzuolanic 
 activity. Thus its tendency to unite with lime liberated by 
 the setting of Portland cement would be likely to be small. 
 It may be regarded for practical purposes as a diluent and 
 makeweight. But this is not all; for much of this slag 
 contains a notable proportion of sulphur, and the objection- 
 able effect mentioned above, of the slow oxidation of this 
 sulphur, holds equally in this case. 
 
 " It is evident tkat the addition to Portland cement of 
 blast-furnace slag, as usually practised, is not only an 
 adulteration, but it is also an adulteration with a dangerous 
 ingredient. Fortunately, this form of fraud is readily 
 detected by analysis, although often escaping recognition 
 by the usual mechanical tests. 
 
 " In conclusion, we may reiterate our views as clearly and 
 briefly as possible. 
 
 "All materials added to Portland cement after the clinker 
 comes from the kilns are adulterants, with the exception of 
 gypsum, which is a recognised addition for a specific 
 purpose in quantities not exceeding 2 per cent. Of the two 
 adulterants which have been specially dealt with, viz., rag- 
 stone and blast-furnace slag, the latter is by far the more 
 objectionable, and it should be condemned and rejected by 
 makers and users alike. In this view we believe we are 
 
 K 2 
 
132 
 
 Tensile Strength in kilos per square inch. 
 
 
 
 
 
 1 
 
 -- 
 
 "t/5 
 
 
 ^j 
 
 13 
 
 "d 
 
 .-( 
 
 s 
 
 
 
 
 c 
 
 
 
 
 1 
 
 . 
 
 % 
 
 *o^ 
 
 S-d 
 
 
 o 
 
 C ^3 
 <U C 
 
 G . "rt 
 <U w C 
 
 
 
 y ^ 
 
 Cu rt 
 
 
 <u 
 
 fi 
 
 s d 
 
 . ^ 
 
 " ^3 
 
 
 J-l 
 
 <u o 
 
 <u o 
 
 S-Sl 
 
 s- a l 
 
 
 
 C4 O 
 
 HMO 
 
 1 1 
 
 11 
 
 
 
 
 
 H H VO 
 
 M MO 
 
 
 Water, 
 
 Water, 
 
 Water, 
 
 Water, 
 
 Water, 
 
 
 25 
 
 8* 
 
 10 
 
 10* 
 
 u^ 
 
 
 per cent. 
 
 per cent. 
 
 per cent. 
 
 per cent. 
 
 per cent. 
 
 
 ' 7 days .. .. 
 
 283 
 
 123 
 
 75 
 
 79 
 
 1 60 
 
 bo 
 
 i month 
 
 359 
 
 153 
 
 119 
 
 M3 
 
 453 
 
 1 
 
 3 months 
 
 447 
 
 207 
 
 163 
 
 202 
 
 294 
 
 T2 
 
 6 months 
 
 
 
 235 
 
 197 
 
 237 
 
 330 
 
 w 
 
 12 months 
 
 
 
 265 
 
 222 
 
 453 
 
 
 
 Water, 
 
 Water, 
 
 Water, 
 
 Water, 
 
 Water, 
 
 
 23* 
 
 12 
 
 12* 
 
 12* 
 
 13* 
 
 
 per cent. 
 
 per cent. 
 
 3er cent. 
 
 per cent. 
 
 per cent. 
 
 
 7 days . . 1 1 " 
 
 485 
 
 I2 5 
 124 
 
 58 
 67 
 
 47 
 74 
 
 260 
 255 
 
 1 
 
 i month I c " 
 ( o . . 
 
 496 
 611 
 
 177 
 206 
 
 185 
 
 139 
 217 
 
 357 
 414 
 
 G 
 
 3 months jo" 
 
 635 
 720 
 
 259 
 286 
 
 194 
 339 
 
 218 
 363 
 
 428 
 469 
 
 ^ 
 
 6 months j c " 
 
 (, o . . 
 
 638 
 
 824 
 
 304 
 310 
 
 232 
 
 407 
 
 347 
 419 
 
 459 
 490 
 
 
 j.2 months 1 g " 
 
 638 
 783 
 
 347 
 366 
 
 295 
 421 
 
 399 
 428 
 
 536 
 524 
 
 N.B. F = fresh water; S = salt water, containing CaSO 4 
 1-0214; MgSO 4 , 0-238; NaCl, 0-1515 grm. per litrs. 
 
 f SiO 2 , 59-04; Al 2 O 3 + Fe 2 O 3 , 14-66; CaO, 1-38; SO 3 , 2 -66; 
 loss on calcination, 21 -76 per cent. 
 
133 
 
 supported by the great majority of engineers and manufac- 
 turers." 
 
 Erdmenger a v nd Lundteigen* have published the tables on 
 page 130 showing the tensile strength in parallel experiments 
 of cements with various admixtures with both fresh and salt 
 water. 
 
 * Thoiiind. Zeit., 22 [Si] , 929-930. Journal, Soc. C.I., vol. xvii., 
 p. 1049. 
 
'34 
 
 CHAPTER XL 
 
 THE STRENGTH OF BRICKWORK. 
 
 THE most recent information as to the strength of 
 brickwork is that contained in a series of three most 
 valuable reports presented by the Science Standing Com- 
 mittee, to the Institute of British Architects during the years 
 18967. The student who desires to study these in detail 
 will find much valuable data for digestion therein, as well as 
 in the discussion of the reports by the members of the 
 Institute. For the present purpose it will suffice to give the 
 salient features and general outcome of the long series of tests 
 which were conducted by a most able Sub-Committee, con- 
 sisting of Messrs. Burrows, Street, Unwin, Max Clarke, 
 Garbutt, and Bernard Dicksee ; the principal part in the 
 experiments being undertaken by Messrs. Street, Clarke, 
 Garbutt, and Bernard Dicksee. The experiments were con- 
 fined to the following varieties of bricks, as affording types 
 of the several kinds most generally employed, viz. : (i) 
 London stocks. (2) Gault. (3) Leicester red. (4) Staf- 
 fordshire blue. In the third series of tests Fletton bricks. 
 Two piers of each kind, 6ft. high and i8in. square, were 
 built, both with lime and cement mortar, for crushing at the 
 end of three months, and a similar number for crushing at 
 the end of nine months. 
 
 The reason for deciding upon having tests at two 
 different periods was to ascertain what additional strength 
 the brickwork gained in six months, as this is of im- 
 portance, considering the great rapidity with which brick 
 buildings are now run up, and sometimes loaded with great 
 weights while the brickwork is quite green, and very little of 
 the mortar set. 
 
The experiments were conducted upon a vacant piece of 
 ground in front of the engineer's office at the West India 
 Docks, and close to the hydraulic engine-house, from which 
 water at a pressure of 700 Ib. to the square inch was avail- 
 -able ; a very powerful hydraulic press, together with a large 
 and massive testing machine, placed at the disposal of the 
 Committee by Sir William Arrol. 
 
 The work of building the experimental piers commenced 
 on the 24th July, 1895, and was finished on the loth; 
 August. They were built upon a temporary line of rails 
 of about i4oft. in length. In the centre of this line of 
 rails Sir William Arrol's testing machine was erected. The 
 piers were built upon wrought iron plates 2ft. by 2ft. 6in. 
 by Jin., placed upon the rails ift. apart, and having two 
 holes drilled in the end nearest the testing machine, by 
 means of which they were drawn along the rails to the 
 machine. 
 
 The bricks, lime, and sand were such as would be 
 supplied to any builder in the ordinary way, while the Port- 
 land cement was of ordinary quality. The cement mortar was 
 mixed by measure in the proportion of i of cement to 4 of 
 washed river sand, and the lime mortar i to 2. The 
 brickwork started upon a bed of mortar of the thickness of 
 an ordinary joint, and rose, as nearly as possible, four 
 courses to the foot. The bond was the same as that shown 
 in Rivington's text -book, and was the ordinary way of build- 
 ing an i Sin. pier. 
 
 As showing the great difficulty of controlling the 
 operations of the men employed in the work, the Com- 
 mittee stated with regret, that during the absence of members 
 of the Committee, the bricklayers, finding the Leicester red 
 and the Staffordshire blue bricks very hard to cut, filled in 
 a great portion of two piers with closers of London stocks- 
 
This quite destroyed the value of these piers, and put the- 
 Committee to the expense of building fresh piers to take 
 their place. This unintentional outcome of part of the ex- 
 periments is, however, not without its value, as indicating 
 the care which should be exercised in controlling the work of 
 the men employed. 
 
 The following extracts from notes by Mr. Matt. Garbutt 
 on the behaviour while under compression, of the first 
 series of brickwork piers, will indicate the character of 
 the results obtained. 
 
 The observed effects of pressure were similar in most 
 cases, but varied in degree with the different materials.. 
 Generally, the first evidences of a strain were slight internal 
 crackling sounds, audible only on applying the ear to the- 
 brickwork. Then the mortar squeezed out of the joints- 
 and fell off this occurring to a greater extent with the- 
 harder bricks, and being more marked where lime mortar was. 
 used than with cement. This was accompanied or followed 
 by cracks occurring in single bricks, which, from slightly 
 irregular bedding or other causes, were unevenly stressed, 
 and by the spalling off of small corners or pieces of the face 
 for similar reasons. By this time the joints were, generally, 
 noticeably compressed, and serious cracks, indicating final 
 lines of rupture, appeared. The piers at this point gener- 
 ally bulged outwards slowly at first, and then, the resist- 
 ance of the brickwork ceasing, the falling of the pressure 
 gauge gave warning that actual collapse was imminent. In 
 the case of the harder bricks, set in cement, the final failure 
 was much more sudden than in the others, so much so that 
 some agility was necessary on the part of the man who, to- 
 measure its compression, stood close to the pier. 
 
 From the beginning of the tests, all through the series, it 
 was evident that the vertical line of joints formed by the- 
 
137 
 
 closers was a plane of weakness, and it was generally at this 
 line that the serious cracks first showed themselves. The 
 majority of the piers broke by crushing in the normal way, 
 or in a manner closely approximating thereto. The pier, 
 No. 3, of Gault bricks in mortar, presented a good typical 
 example. It bulged evenly on all sides at once, nearly 
 every brick in the body of the pier was broken, and the two 
 
 First Series. 
 
 
 
 
 Press 
 
 ure per 
 
 
 
 
 Age of 
 
 sq. ft. ( 
 
 )f pier, in 
 
 
 Brick. 
 
 ^sed^ 
 
 pier. 
 
 tons, i 
 the 
 
 it which 
 5 pier 
 
 Average. 
 
 
 
 Weeks. 
 
 began 
 to fail. 
 
 collapsed. 
 
 
 London stock 
 
 Lime 
 
 18-3 
 
 4-18 
 
 10-41 
 
 10-41 
 
 Gault 
 
 , 
 
 18-3 
 
 5'oo 
 
 21-82) 
 
 
 M 
 
 
 18-8 
 
 6-16 
 
 22-031" 
 
 21 "92 
 
 Leicester red. . 
 
 
 19-1 
 
 15-20 
 
 29-93! 
 
 
 
 
 
 18-6 
 
 
 J C - 1 
 
 30-74 
 
 Staffordshire blue.. 
 
 n 
 
 
 
 18-6 
 19-4 
 
 22-43 
 21-42 
 
 69-22) 
 
 79 '39/ 
 
 74-30 
 
 London stocks 
 
 Cement. 
 
 2I'I 
 
 7-22 
 
 16-03) 
 
 
 ,, 
 
 t 
 
 21 -0 
 
 
 13-83) 
 
 14-93 
 
 Gault 
 
 , 
 
 21 -O 
 
 6-98 
 
 18-07) 
 
 
 ?) 
 
 ( 
 
 20 -5 
 
 7-08 
 
 I7-5I)" 
 
 17-79 
 
 Leicester red . . 
 
 , 
 
 22T 
 
 17-87 
 
 67-36) 
 
 
 ,, 
 
 ( 
 
 21 *7 
 
 21-82 
 
 49 ' 54 / 
 
 5 8 45 
 
 Staffordshire blue.. 
 
 , 
 
 22-7 
 
 16*91 
 
 84-47\ 
 
 72-80. 
 
 
 
 
 
 
 
 pyramids at top and bottom, remaining after the crushing 
 of the mass, were regularly defined and approximately con- 
 centric with the axis of the pier. 
 
 In one case the cement was found to be disintegrated, 
 and as the corresponding pier, which was subjected to 
 almost exactly the same pressure, showed its cement to be 
 perfectly hard set, and uninjured by crushing, one is led to- 
 
-the conjecture (says Mr. Garbutt; that the cement used in 
 No. 12 had been left mixed overnight and beaten up again 
 in the morning, or that the bricks had not been sufficiently 
 wetted, and had withdrawn the moisture from the cement. 
 
 The results of the tests of the first series, which are set 
 out in the reports in very full detail, may be shortly sum- 
 marised as in the table on the preceding page. 
 
 The average strengths of the bricks used in building the 
 above piers, as ascertained by Professor Unwin by his tests 
 at the Central Technical College, are as follows : 
 
 Crushing strength 
 in tons per square 
 foot. 
 
 London stocks 84-27 
 
 Gault 182-2 
 
 Leicester red 382-1 
 
 Staffordshire blue 701 i 
 
 The results of similar tests of lime and cement mortar 
 made in the proportions used gave the following : 
 
 Lime Briquettes. 
 Strength at 4 weeks .. .. 6-08 tons per sq. ft. 
 
 ,, 12 8'73 
 
 .. 24 , 15-72 
 
 Cement Briquettes. 
 
 Made with same sand as that used in building the piers. Strength 
 at 24 weeks = 29-00 tons per sq. ft. 
 
 Made with Leighton Buzzard sand. 
 
 Mean strength at 4 weeks . . . . 31 '45 tons per sq. ft. 
 13-7 4 8 '52 ,, 
 24-0 , 56-15 
 
 In order to ascertain the composition of the mortars used 
 :in the above experiments, a series of analyses were made in 
 
139 
 
 the laboratory of the London County Council under my 
 direction, at the request of the Committee, conveyed 
 through Mr. Blashill, at that time superintending architect 
 of the Council, with the results given in table on next page. 
 
 When reporting upon the second series of tests, the 
 Committee observed that it was interesting to note the great 
 difference between the strength of bricks and of brickwork, 
 the enormous disproportion being due to the quality of the- 
 material interwoven with the bricks as a means of uniting 
 and holding the mass together. The effect upon different 
 kinds of bricks was variable, owing to some bricks being more 
 susceptible to adhesion than others. Again, if the mortar 
 be imperfectly mixed, or the joints and beds of the brick- 
 work imperfectly filled up, the injurious effect on strength 
 is incalculable. It introduces unequal pressures in different 
 parts of the body, and the work fails in detail, until so 
 much of the whole is destroyed that the remainder suddenly 
 collapses. 
 
 The Committee laid much stress upon the care required 
 in the selection, preparation, and application of the mortar. 
 They observed, however, "that there is one comforting reflec- 
 tion, namely, that if a building is likely to collapse from 
 want of strength, it is more likely to do so while the mortar 
 is fresh and green than at a later date ; that is to say, if the 
 mortar once gets set and matured under its load, the build- 
 ing will live its life, a longer or a shorter one, according to 
 the character of the building" or, shall one say, of the 
 builders ? A few weeks ago, observes the Committee, in 
 taking down some buildings less than a hundred years old, 
 the bricks could be lifted piecemeal by the hand, and an 
 overturn of 3ft. or 4ft. of the walls was sure to dislocate 
 the whole of the bricks. On the other hand, in some 
 buildings several centuries old, the mortar joints are as hard 
 
140 
 
 c 
 
 o 
 
 <U T3 C 
 
 
 
 *J 
 
 
 
 
 
 
 
 
 
 sJ^ 
 
 S 
 
 ri 
 
 N 
 
 * 
 
 N 
 
 n- 
 
 M O 
 
 in 
 
 M 
 
 R 
 
 ^- 0? 
 
 8 
 
 <u 
 
 S 
 
 ^co^ 
 
 
 
 u 
 
 I 
 
 r 
 
 M 
 M 
 
 H 
 
 O "f 
 
 N 
 
 
 
 04 M t^ 
 
 8 
 
 ^ 
 
 IH 
 
 <U 
 
 S 
 
 ffi 
 
 M^CO^ 
 
 cement 
 
 mortar. 
 
 Per cent. 
 6-16 
 
 8 
 
 M 
 
 <f IO 
 
 co o 
 b V 
 
 8 
 
 01 
 
 in 
 
 oo 
 
 b 
 
 o 2 to 
 
 VN MO 
 
 8 
 8 
 
 6 
 5? 
 
 V-i 
 
 <u 
 
 .^'S^ 
 
 >" 
 
 J - 
 
 M oo *2 
 
 cement 
 
 _; 
 
 a 
 
 S 
 
 er cent. 
 
 5'72 
 
 IN 
 
 H 
 
 C 
 
 o> o 
 o o 
 
 o ^t* 
 
 rr> 
 
 P 
 c 
 
 ^. 8 
 
 ^ 
 
 8 
 8 
 
 UH 
 
 
 .S 
 
 ~ 
 
 PH 
 
 
 
 
 
 
 
 M 
 
 | 
 
 a .9 
 
 1 
 
 a 
 
 o ^ 
 
 cc~ 
 
 VO 
 
 o m 
 
 
 
 o" 
 
 ^ oo t^ 
 
 8 
 
 K 
 
 -&I 
 
 8 
 
 i 
 
 JH CO 
 
 
 
 M 
 
 01 
 
 o m 
 
 01 
 
 o 
 
 M NO 
 
 8 
 
 IH 
 
 <U 
 
 OH 
 
 O Q .2 
 
 to 
 
 cement 
 
 mortar. 
 
 o p 
 
 r 
 
 eg 
 
 01 
 
 m 
 CS fO 
 
 b V 
 
 CO 
 
 o 
 
 o >n ^ 
 
 O IT) CO 
 
 8 
 
 2 
 
 6 
 
 'S 
 
 
 B 
 
 g^ 
 
 
 
 t*>> in 
 
 
 
 CTv ro t^ 
 
 
 " 
 
 3 v2 
 
 Z- 
 
 .5 
 
 o 
 
 vO 
 
 c 
 
 M O 
 
 *n 
 
 00 
 
 o on- 
 
 
 _E 
 
 ^s'S 
 
 
 
 1 
 
 r 
 
 O 
 
 M 
 
 w 
 
 O M 
 
 o 
 
 
 
 - R. 
 
 o 
 
 | 
 
 _fe i 
 ^ "55 
 
 
 
 S 
 
 |a 
 
 CO 
 M 
 
 % 
 
 M 
 
 m oo 
 
 8 
 
 S 
 
 r^ M i-i 
 
 O M CO 
 
 8 
 
 
 in o -* 
 
 
 
 -i O 
 
 <N 
 
 M 
 
 N 
 
 H 
 
 
 
 M N M 
 
 
 
 
 S^ 
 
 3 
 
 
 
 & 
 
 
 
 
 
 
 ^** 
 
 2 
 
 6 
 ^ 
 
 ^^" 
 
 
 
 y 
 
 ^ 
 
 0^ 
 
 CJ 
 
 
 o 
 
 * 
 
 *3- in to 
 
 t^> CO VO 
 
 R 
 
 
 
 E 
 
 
 
 
 
 
 
 
 
 
 O) 
 
 SD % 
 
 
 
 
 <u 
 
 m 
 
 M 
 
 IN 
 
 o 
 
 c 
 
 M N O^ 
 
 8 
 
 ffi 
 
 
 
 c 
 
 p l 
 
 
 
 
 
 
 
 2 
 
 
 
 
 
 
 
 
 
 
 
 
 . c . 
 
 
 
 
 
 
 
 
 
 . 
 
 
 
 . 
 
 . 
 
 rt 
 
 . 
 
 
 
 
 
 
 
 <D 
 
 c 
 
 
 
 G ^B 
 
 
 
 
 
 
 ; 
 
 ; 
 
 rt 
 
 1 * 
 
 rt 
 
 o 
 
 g 
 
 O X 
 
 ^ HH ^_ 
 
 g C 
 
 ; 
 
 
 
 
 
 
 6 
 
 .Q 
 
 Q-. ^ 
 
 ~ 
 
 "rt 
 
 
 
 
 
 
 
 o 
 
 rt 
 
 rt 
 
 
 CO 
 
 g 
 
 HH o : -a 
 
 "rt 
 
 
 
 
 
 2 
 
 4} 
 
 O 
 
 CO C 
 rt 
 
 ^ 
 
 ^ 
 
 a3c rt 
 
 (2 
 
 
 
 
 
 "o 
 
 S 
 
 ; 
 
 C 
 
 ~ 
 
 z 
 
 'c 
 
 VH 
 
 g rt C 
 
 
 
 
 
 
 !; 
 
 i-4 
 
 
 M 
 
 CO 
 
 w 
 
 n4 ^ CO 
 
 
First and Second Series, 
 
 Brick. 
 
 Mortar 
 used. 
 
 Age of pier. 
 
 Pressure 
 per sq. ft. 
 in tons at 
 which 
 the pier 
 collapsed. 
 
 Average. 
 
 Stocks 
 
 Lime 
 
 3 to 4 months 
 
 10-41-) 
 
 
 Jf 
 
 M 
 
 M 
 
 *i8-28/ 
 
 X 4"34 
 
 
 
 i) 
 
 10 months 
 
 14-111 
 
 I2 '54 
 
 Gault 
 
 
 3 to 4 months 
 
 21-82-) 
 
 
 
 M 
 
 
 22-03; 
 
 21-92 
 
 
 
 " 
 
 10 months 
 
 20-27; 
 
 21-60 
 
 Leicester red. . 
 
 
 
 3 to 4 months 
 
 29-93) 
 3I-55; 
 
 30-74 
 
 .. .. 
 
 " 
 
 10 months 
 
 34-35) 
 33-89; 
 
 34-12 
 
 Staffordshire blue . . 
 
 
 
 3 to 4 months 
 
 69-22-) 
 
 
 jj * 
 
 M 
 
 
 79'39J 
 
 74'3 
 
 ji 
 
 p 
 
 10 months 
 
 
 
 
 
 
 73 ' ?6 1 
 
 73-66 
 
 " 
 
 1 > 
 
 11 
 
 73 7 J 
 
 
 Stocks 
 
 Cement 
 
 3 to 4 months 
 
 16-03) 
 13-83' 
 
 14 '93 
 
 >t 
 
 t( 
 
 10 months 
 
 22-34) 
 
 
 }> 
 
 >f 
 
 
 16-96; 
 
 19-65 
 
 Gault 
 
 
 
 3 to 4 months 
 
 18-0/1 
 
 
 
 
 
 
 " 
 
 %8;ioj 
 
 33-68 
 
 
 
 " 
 
 10 months 
 
 II 
 
 26-331 
 33-64J 
 
 29-98 
 
 Leicester red . . 
 
 ,, 
 
 3 to 4 months 
 
 67-36) 
 
 
 " .... 
 
 " 
 
 " 
 
 /9'54r 
 
 67-76 
 
 j> .... 
 
 ,, 
 
 10 months 
 
 47-86) 
 
 
 Staffordshire blue.. 
 
 " 
 
 3 to 4 months 
 
 53'00/ 
 
 5 '43 
 
 ii 
 
 ji 
 
 
 61-14! 
 
 
 u 
 
 
 
 
 87-94 
 
 j 
 
 M 
 
 " 
 
 *97'9J 
 
 
 " 
 
 " 
 
 10 months 
 
 91 -Sen 
 
 82-48 
 
 Leicester red. . 
 
 In sand 
 
 
 
 15-0 
 
 15-0 
 
 ! only 
 
 
 
 
 * The piers were built in a manner very superior to the first two 
 series, and therefore cannot fairly be averaged with them. 
 
142 
 
 " as iron," the joints protruding beyond the masonry, the 
 "surface of which had fretted away by the action of the 
 atmosphere during the ages which had elapsed since the 
 structure was built. 
 
 The results of the first and second series of tests are given 
 in the table on page 141. 
 
 It will be readily noticed, observes the Committee, in 
 looking down the columns, that something must have 
 greatly disturbed these averages, as it would, for instance, 
 be absurd to expect some of the descriptions to be weaker 
 at ten months than at four. The cause, however, is not far 
 to seek. The piers built as substitutes for those which the 
 Committee found badly built in the first series, were so 
 superior to any of those reserved for the ten months' test, 
 that they raised the average for four months to beyond that 
 obtained for the piers crushed at ten months. Taking, 
 therefore, the original piers as specimens of what might be 
 got in ordinary practice, one ought, perhaps, to reduce the 
 averages to approximate to those obtained in the first in- 
 stance from indifferent, which is indeed, the ordinary manner 
 of building. 
 
 In the third report the Committee gave the results of ex- 
 periments conducted to ascertain the average strength of 
 various descriptions of brickwork by crushing twenty short 
 lengths of brick walls, each about 6ft. high by 2yin. long 
 and i Sin. thick. The results are set out, as before, in con- 
 densed tabular form on page 143. 
 
 Pending a more careful examination of the facts, the Com- 
 mittee derived the following impressions. The resistance 
 of brickwork in lime mortar to crushing would seem to vary 
 from one-sixth to one-eighth of the resistance offered by the 
 brick itself, while in cement mortar it varies from one-half 
 to one-fifth of that strength. It is obvious that while 
 
cement mortar must very materially aid the weaker bricks in 
 their combined strength, it cannot materially affect the 
 ultimate powejr of resistance in brickwork made of a 
 harder variety. 
 
 The average thickness of the bricks was 2|in., and" the 
 
 Third Series Brick Walls. 
 
 
 
 Pressur 
 
 e 
 
 
 Brick. Mortar 
 
 Age of wall. 
 
 per sq. i 
 of wall i 
 tons al 
 
 t. 
 n 
 
 Average. 
 
 
 
 which w 
 
 all 
 
 
 
 
 collapse 
 
 J. 
 
 
 Stock . . . . Lime 
 Gault .. .. ! 
 
 22 weeks 
 
 I9-83J 
 
 
 18-63 
 
 Fletton .. 
 
 . 
 
 30-94 
 30-82 
 
 
 30-68 
 
 Leicester red . . % , 
 Staffordshire blue . . 
 
 
 
 45 "94 
 44-78 
 118-12 
 110-56 
 
 \ 
 
 U4-34 
 
 Stocks . . . . Cement 
 Gault .. .. 
 
 21 weeks 
 22 weeks 
 
 39-24! 
 39-34. 
 
 ' 
 
 5 I- 34 
 
 Fletton .. .. !, 
 
 21 weeks 
 
 54-88 
 
 
 56-25 
 
 Leicester red .. ,, 
 Staffordshire blue . . ,, 
 
 
 
 80-94 
 85-78 
 139-52 
 
 
 I35-43 
 
 total thickness of the mortar beds was 6in., while the com- 
 pression of the lime mortar beds averaged iin., and that of 
 the cement mortar beds about ^in. This proves that the 
 mortar generally was well crushed, and disintegrated long, 
 before the final collapse of the several examples of brick- 
 
144 
 
 work. The instantaneous photographs taken by the Com- 
 mittee showed the mortar flowing out as in a stream or 
 fountain at the moment of collapse. 
 
 In dealing \\ith the working load that may be calculated 
 upon, care must be taken not to impose such a load as 
 would materially damage the structure of the brickwork. 
 At one-fifth of the crushing 'load the compression in lime 
 mortar averaged -^in. in 6ft. of brickwork, and in cement 
 mortar it averaged -g^in. 
 
 Another thing that will have to be remembered is the 
 great difference between dead and live loads. 
 
 If, says the Committee, we take the safe load, or one that 
 would not materially damage the structure, as one-fifth of 
 the crushing load, it may be assumed from the results 
 obtained, that in lime mortar i to 2, stock brickwork is equal 
 to about 3^ tons ; Gault, 6 tons ; Fletton, 6 tons ; Leicester 
 red, 9 tons ; and Staffordshire blue, 23 tons per square foot. 
 In Portland cement mortar, i to 4 stocks would be equal to 
 .about 8 tons; Gault, 10 tons; Fletton, n tons; Leicester 
 red, 17 tons; and Staffordshire blue, 24 tons per square 
 foot. This, however, is put forth as only a general assump- 
 tion, which requires further consideration. 
 
'45 
 
 CHAPTER XII. 
 
 CONCRETE. 
 
 Definition. Concrete is a solid mass formed from lime 
 or cement, with sand, and small, irregular pieces of stone, 
 brick, slag, clinker, &c., by their amalgamation together 
 when brought into contact with each other by the aid of 
 water. The amalgamative agents, or more properly speaking, 
 the solidifying agents, are the lime or cement and the water; 
 the other ingredients are inactive, being merely bound 
 together by the foregoing. The two classes of ingredients 
 which go to form concrete are technically known as the 
 matrix and aggregate, i.e., the active (cement and water) and 
 the inactive agent (sand) respectively. 
 
 Beton. Concrete is sometimes called by the French 
 name of Beton. Between these two names there is a dis- 
 tinction without a difference, for both signify " concrete " 
 as the word is understood in its wide sense. The distinc- 
 tion, however, lies solely in the operation of mixing the 
 ingredients as practised in England and France. In this 
 country the general practice is to mix the whole of the 
 ingredients dry, and when they are thoroughly mixed water 
 is added gradually, taking care, however, to turn over the 
 ingredients during the addition of water. In France it is 
 customary, first, to form a mortar by mixing the lime or 
 cement with the sand and water ; the bricks, stone, or other 
 material being added to the mortar, thus formed, after- 
 wards. 
 
 Historical. If, in considering the history of concrete, we 
 include that of mortar and cement, we must go back very 
 
146 
 
 far indeed to arrive at the time when this substance in one- 
 or more of its various modifications was first employed in 
 building construction. In the opinion of some authorities, 
 the pyramid of Cheops is an excellent example of how little 
 we have advanced in this respect on the knowledge of the 
 ancient Egyptians; who, when they required the early Israel- 
 ites to make "bricks without straw," evidently wanted a crude 
 sun-dried "mortar" or "concrete" in blocks. It can hardly 
 be contended, however, that these early bricks come under 
 the head of mortar or concrete, but yet they point to a means- 
 of preparing earthy matters for building purposes, and 
 doubtless the introduction of ashes and, later, lime, to the 
 clay and sand, gradually developed the present system of 
 making concrete as we now know it, with Portland cement 
 and ballast. In Prescott's " Conquest of Mexico," stucco 
 is frequently referred to, as also, "the sensations of the 
 Aztecs as they heard for the first time, the well-cemented' 
 pavement ring under the iron tramp of the horses." It is 
 certain that the Etruscans and Greeks imparted their know- 
 ledge in this respect to the Romans, who employed con- 
 crete in the construction of Corfe Castle in Dorsetshire, &c. 
 The use of Roman cement may be seen in various ancient 
 buildings at Pevensey and Richmond, in the Norman and 
 early English foundations of Ely Cathedral, in Salisbury 
 Cathedral, Westminster Abbey, Guildford Castle, Middle- 
 ham Castle (Yorkshire), and many other old English 
 castles. An inspection of the ruins of feudal fortifications 
 reveals the fact that concrete was very commonly used in 
 the Middle Ages. The Castle of Badajos, in Spain, still 
 bears the marks of the boarded frames in which the 
 concrete was deposited. To the Chinese we must give 
 credit for having been, if not the first, amongst the earliest 
 nations who used concrete. The Great Wall of China, begun. 
 
M7 
 
 in 214 B.C., was built mainly of concrete, but a "concrete" 
 different to both our own method of mixing the ingredients 
 and that of the v French method. 
 
 Chinese Method of Concrete Making. The Chinese 
 method of preparing concrete for wall building is- as 
 follows : The ingredients are i part lime and 2 parts of a 
 mixture of either sand, gravel, and shingle, or sand, building 
 debris, and earth. These ingredients are mixed thoroughly 
 together dry. The wall intended to be built is first outlined 
 by a wooden shell, or rather by boarded frames, and filled 
 up by degrees with the dry concrete mixture described 
 above. A layer of this mixture, about 6in. deep, having 
 been put into the casing or shell of the intended wall, the 
 workmen jump into the casing and begin to press and 
 pound the mixture down by means of rammers wooden 
 poles having a base of about 6in. to Sin. diameter, tapering 
 towards the top. The concrete mixture is by this means 
 compressed to a sufficient compactness. Over this punned 
 or rammed surface water is sprinkled, by means of a wooden 
 bowl dexterously used by the ioremen, whereby the surface 
 is moistened, but not wetted ; then some more of the loose 
 concrete mixture is put into the casing, and the processes 
 are repeated until the intended height of the wall is reached. 
 A wall built in this manner divides the Portuguese city of 
 Macao, China, from the Chinese territory of Heong-San. 
 It has stood for nearly 250 years, and my informant, Mr. 
 Francombe, who has lately seen it, says, " It is still in 
 splendid condition." 
 
 The Romans were great concrete builders. They 
 made use of it before the year 500 B.C. Vitruvius, who- 
 wrote circa 100 B.C., gives instructions for its manufacture. 
 He also describes the process of making concrete for 
 pavements and floors, and concrete walls around wells. 
 
 L 2 
 
148 
 
 His specification lor the concrete in the last instance is as 
 follows (according to Gwilt's translation) : " In the first 
 place, the purest and roughest sand that can be had is to 
 be procured. Then material is to be prepared of broken 
 flint, whereof no single piece is to weigh more than i Ib. 
 The lime must be very strong, and in making it into 
 mortar five parts of sand are to be added to two of lime. 
 The flint work is combined with the mortar, and of it the 
 walls in the excavation are brought up from the bottom, and 
 shaped (Professor Aitchison, A.R.A., renders the word 
 " rammed " instead of " shaped ") by wooden bars covered 
 with iron." 
 
 Alberti and Palladio, in the fifteenth and sixteenth 
 centuries, described methods of building concrete walls. 
 Amcngst the works of note constructed in more recent 
 dates wherein concrete was used may be mentioned the 
 Millbank Penitentiary (in 1811), by Sir Robert Smirke; 
 the Graving Dock and Sea Wall at Woolwich (1835), by 
 Ranger. 
 
 Concrete in Building Construction. The employment of 
 concrete in building construction has received a great 
 impetus since the introduction of Portland cement un- 
 doubtedly the strongest cement the world has ever known 
 which has on account of its great strength and reliability led 
 to the use of concrete for many purposes for which lime- 
 concrete was utterly unsuitable, with the result that concrete 
 is now used in foundations, floors, walls, roofs, drain pipes, 
 stairs, conduits, arches, pavings, bridges, lintels, building- 
 blocks, and has even been employed in making doors, 
 telegraph poles, and in the erection of mill chimneys, either 
 alone or in combination with wood, iron, steel, c. With 
 the advancement made in the use {of cement-concrete, 
 numerous patents have been taken out for improvements 
 
149 
 
 both in the methods of making and using this substance, 
 some of varying merits, while others have been failures 
 involving great* pecuniary losses to the inventors; and to- 
 day concrete, notwithstanding many failures the stepping- 
 stones to success or advancement is used for a larger 
 variety of purposes than ever before. 
 
 Mixing. The nature of limes and cements for specific 
 concrete works have been dealt with in a previous chapter. 
 In the preparation of concrete a great deal depends upon 
 how the matrix is mixed with the aggregate ; its importance 
 has been recognised by the early users of concrete, with the 
 result that various concrete-mixing machines have been 
 invented some made in sizes small enough to be worked 
 by hand; but those driven by steam power are the ones 
 mostly in general use. Concrete is more thoroughly and 
 rapidly mixed by machinery than by hand, and for consider- 
 able quantities it is much cheaper, apart from the fact that 
 machines do their work with more uniformity than labourers, 
 and constant supervision, so necessary when the latter is 
 employed, may be lessened. Many architects are now 
 specifying that all concrete in large quantities shall be mixed 
 by machinery. Care will, however, have to be experienced 
 by them in the choice and approval of concrete-mixing 
 machines, as many of these seem designed rather to turn 
 out a large quantity of concrete than to ensure the thorough 
 mixing of the materials. The thorough incorporation of the 
 matrix with the aggregate should be the main point to be 
 considered in the selection of these machines. 
 
 Depositing. As soon as concrete is mixed, no time 
 should be lost in depositing it, as any disturbance of the 
 mass after the cement or lime has begun to set detracts 
 from the ultimate strength ; more special attention in this 
 regard should be given where quick-setting cement is 
 
'5 
 
 Concrete Block : Compressed. 
 
 Portland cement concrete blocks of various materials, set and 
 kept in air for one year ; also set and kept in water for the same 
 time. Materials, various; size of block, 6in. by 6in. by 6in.; 
 moulded, November 4th, 1867; tested, November 4th, 1868. 
 
 d 
 o 
 
 1 
 1 
 
 CL 
 6 to i 
 
 Weight in pounds. 
 
 Weight of 
 each block 
 in pounds. 
 
 Crushed at 
 tons. 
 
 Remarks. 
 
 Cement. 
 
 oJ 
 
 in 
 
 rt 
 
 .s 
 
 *-" JH 
 
 .2 
 ^ 
 
 JH* 
 
 Water. 
 
 2-26 
 
 i5-3i 
 
 i -oo 
 
 17-90 
 
 18-60 
 
 20- 4 
 
 19-60 
 
 Ballast 
 
 
 2-40 
 
 12-80 
 
 -45 
 
 16-90 
 
 17-67 
 
 40-60 
 
 34-50 
 
 f Portland 
 \ stone 
 
 
 2-31 
 
 H'S 6 
 
 1-30 18-53 
 
 18-88 
 
 30-50 
 
 27-0 
 
 Granite 
 
 
 2-16 
 
 I3-33 
 
 !'75 
 
 16- 10 
 
 17-50 
 
 28-80 
 
 26*50 
 
 Pottery 
 
 
 2'II 
 
 12-29 
 
 i -60 
 
 15-08 
 
 16-61 
 
 23-0 
 
 23-50 
 
 Slag 
 
 
 2-03 
 
 13-91 
 
 *"45 
 
 16-50 
 
 17-65 
 
 20-50 
 
 24-0 
 
 Flints 
 
 
 2-37 
 
 J5-5 1 
 
 1-30 
 
 18-50 
 
 19-25 
 
 28-0 
 
 2 3 -C 
 
 Glass 
 
 8 to i 
 
 I- 7 6 
 
 15-88 
 
 95 
 
 17-86 
 
 18-90 
 
 I3-50 
 
 I3-50 
 
 Ballast 
 
 
 1-86 
 
 13-26 
 
 1-65 
 
 16-32 
 
 17-50 
 
 33-00 
 
 29-00 
 
 f Portland 
 ^ stone 
 
 
 i-53 
 
 I4-57 
 
 1-25 
 
 18-10 
 
 19-0 
 
 I9-6O 
 
 16*00 
 
 Granite 
 
 
 1-64 
 
 I3-53 
 
 i -60 
 
 16-15 
 
 16-95 
 
 22-00 
 
 23-00 
 
 Pottery 
 
 
 i -60 
 
 12-37 
 
 1-50 
 
 14-20 
 
 15-91 
 
 19-50 
 
 13-40 
 
 Slag 
 
 
 1-56 
 
 14-28 
 
 1-30 
 
 !6'45 
 
 17 '62 
 
 17-50 
 
 2O 'OO 
 
 Flints 
 
 
 1-82 
 
 !5'79 
 
 I'I5 
 
 18-00 
 
 18-93 
 
 18-00 
 
 17-50 
 
 Glass 
 
 10 tO I 
 
 1-42 16-11 
 
 8 5 
 
 17-68 
 
 18-70 
 
 10-50 
 
 10-50 
 
 Ballast 
 
 
 1-52 
 
 -3-56 
 
 1-86 
 
 16-44 
 
 17-90 
 
 22 'GO 
 
 16-50 
 
 (Portland 
 ( stone 
 
 
 1-23 
 
 14-60 
 
 I-I5 
 
 17-60 
 
 18-50 
 
 I5-5 
 
 12-40 
 
 Granite 
 
 
 1-28 
 
 12-46 
 
 1-40 
 
 i6'og 
 14-00 
 
 16-80 
 15-50 
 
 18-50 
 10-50 
 
 19-00 
 
 lO'OO 
 
 Pottery 
 Slag 
 
 
 1-26 
 
 14-37 
 
 i-i5 
 
 16-15 
 
 17-60 
 
 15-00 
 
 18-50 
 
 Flints 
 
 
 1-48 : 16-16 
 
 1-05 17-90 
 
 18-73 
 
 12-50 
 
 12 "80 
 
 Glass 
 
Concrete Block : Not Compressed. 
 
 i 
 
 i 
 
 U< 
 
 cu 
 
 Weight in pounds. 
 
 Weight of 
 each block 
 in pounds. 
 
 Crushed at j 
 tons. 
 
 Remarks. 
 Ballaet 
 
 Cement. 
 
 -d 
 J 
 
 Water. 
 
 a 
 
 
 
 J-1 
 
 <- 
 
 w. 
 
 ^g 
 W * 
 
 < 
 
 i 
 
 <C to i 
 
 2'OO 
 
 13*49 
 
 88 
 
 17-35 
 
 18-30 
 
 18-20 
 
 17-00 
 
 
 2*23 
 
 11-91 
 
 i-35 
 
 15-70 
 
 17-12 
 
 30-00 
 
 23-00 
 
 I Portland 
 I stone 
 
 
 1-98 
 
 I2'55 
 
 I'I2 
 
 17-40 
 
 18-30 
 
 24-50 
 
 15-50 
 
 Granite 
 
 
 1-91 
 
 11-78 
 
 1-54 
 
 15-85 
 
 17-00 
 
 24-60 
 
 24-00 
 
 Pottery 
 
 
 i'95 
 
 11-36 
 
 1-23 
 
 14-80 
 
 I5-77 
 
 20 -oo 
 
 19-20 
 
 Slag 
 
 
 1-76 
 
 I2'II 
 
 1-26 
 
 15-20 
 
 16-60 
 
 15-50 
 
 15-50 
 
 Flints 
 
 
 2-04 
 
 I3-35 
 
 I'll 
 
 17-76 
 
 18-40 
 
 16-50 
 
 17-00 
 
 Glass 
 
 S to i 
 
 I-DO 
 
 H'5 1 
 
 *8 5 
 
 17-38 
 
 I7-95 
 
 12-50 
 
 II'OO 
 
 Ballast 
 
 
 I-70 
 
 12" 15 
 
 1-51 
 
 15-63 
 
 17-00 
 
 24-50 
 
 19-50 
 
 f Portland 
 ( stone 
 
 
 I-3I 
 
 12-53 
 
 1-07 
 
 17-45 
 
 18-20 
 
 14.- 50 
 
 13-40 
 
 Granite 
 
 
 i*43 
 
 11-83 
 
 1-40 
 
 I5-7 1 
 
 16-70 
 
 iS'oo 
 
 18-00 
 
 Pottery 
 
 
 i*47 
 
 i-34 
 
 ii-37 
 
 I2'O2 
 
 i-37 
 
 I'I2 
 
 13-73 
 15-40 
 
 15-12 
 16-60 
 
 14-00 
 14*00 
 
 9-50' Slag 
 12*50! Flints 
 
 
 1-52 
 
 I3-22 
 
 9 6 
 
 17-30 
 
 17-80 
 
 13-60 
 
 11-40 
 
 Glass 
 
 10 to i 
 
 1-26 
 
 I4'35 
 
 '75 
 
 17-20 
 
 17-50 
 
 8-00 
 
 7-00 
 
 Ballast 
 
 
 1-38 
 
 12-34 
 
 I-6 9 
 
 16- 15 
 
 17-00 
 
 19 'oo 
 
 lO'OC 
 
 f Portland 
 i \ stone 
 
 
 i-o5 
 
 12-62 
 
 99 
 
 16-97 
 
 17-63 
 
 11-50 
 
 ii *oo Granite 
 
 
 I-I5 
 
 n-88 
 
 I-2 5 
 
 15-66 
 
 16-50 
 
 14-00 
 
 17-00 Pottery 
 
 
 1-17 
 
 "'43 
 
 1-28 
 
 *3*75 
 
 15'oc 
 
 8-50 
 
 5'5 C 
 
 Slag 
 
 
 1-13 
 
 1-22 
 
 I2'00 
 I3-3I 
 
 9 
 
 86 
 
 15-12 
 16-95 
 
 i6"6o 12*80 
 
 17-70 lO'OO 
 
 II'OC 
 
 10 -8c 
 
 i Flints 
 > Glass 
 
employed. When the concrete is in large masses, as in 
 harbour works, caissons, retaining walls, foundations, &c. r 
 it is customary to use packing, that is to say, large pieces 
 of stone are inserted for the sake of economy. This can 
 safely be done provided that all packing stones are thoroughly 
 wetted before being laid in the work, and that they are not 
 too near each other ; care should be taken that there is a, 
 sufficient excess of mortar in order that the packing may be 
 properly united with the rest of the mass. 
 
 Ramming. Compression increases the strength of 
 concrete, as also its density, and, in consequence, 
 its imperviousness to water and its durability. With- 
 out ramming it is impossible to have impervious con- 
 crete. On the other hand, if ramming is resorted to in 
 large masses of concrete, there is the risk of interrupting 
 the process of setting, which commences immediately on 
 the application of moisture. An increased proportion of 
 cement should be added if it is desired to augment the 
 strength of the concrete. Ramming, however, may be 
 safely recommended in making concrete bricks or blocks 
 of moderate size, states my former colleague, the late Mr. 
 John Grant, who carried out a series of experiments to- 
 prove this. In his work on the "Strength of Cement," page 
 140, a series of tables are given, in which he shows the 
 results of his experiments, abstracts from which are repro- 
 duced on pages 150 and 151. 
 
 The compression was effected for the above experiments- 
 by beating the concrete into the moulds with a small mallet. 
 As will be seen by the table given above, the average gain 
 in strength was 
 
 30 per cent, for 6 to i mixture kept in air, 
 37i 6 to i water, 
 
27 J per cent, for 8 to i mixture kept in air, 
 
 39 8 to i water, and 
 
 24 *, 10 to i air, 
 
 44 10 to i water. 
 
 This is a gain that commends itself to builders and to- 
 manufacturers of concrete drain-pipes and artificial stones, as- 
 the latter almost invariably compress the raw material in one . 
 way or another, in order that the goods may be strong and 
 impervious to water. As concrete very frequently has 
 Portland cement as a matrix, it requires protection 
 not only against traffic, but also against extremes of 
 temperature and atmospheric influences, which should 
 be avoided as much as possible. Mr. W. W. Mackay,. 
 of the Docks Department, New York, found in 1876 
 that Portland cement briquettes made and kept in water 
 at a uniform temperature of 60 deg. Fah., were at the 
 end of seven days 30 per cent, stronger than exactly similar 
 briquettes which were made at a temperature of 60 deg. to- 
 70 deg., and kept in water varying daily from 70 deg. down 
 to about 40 deg. Again, Mr. Fitzmaurice gives the 
 results of his experiments in Nova Scotia, during the winter 
 of 1890-1 : "Exposure to natural frost for four days 
 out of a total of twenty-eight days reduced the strength 
 of neat Portland cement briquettes 15 per cent., and of 
 cement and sand briquettes (i to 3) from 28 to 35 per 
 cent. Exposure for the full period of twenty-eight days- 
 reduced the strength of the former 35 per cent, and of the 
 latter no less than 57 per cent." Practical architects 
 maintain that concrete setting at ordinary temperatures is- 
 of satisfactory strength, while it is considerably weakened 
 by frost, and that in extreme cases it is rendered worthless. 
 Hastening the Setting of Concrete. The matrix for making. 
 
154 
 
 such a concrete should be composed of the best quick- 
 setting Portland cement used fresh, and less sand should be 
 used, taking care to reject fine sand and dust, as these 
 greatly retard that action ; and last, but not the least, no 
 more water should be used than that which is absolutely 
 necessary. If a lesser quantity of aggregate be incorporated 
 with this matrix than is usual, the rapidity of the setting 
 of the concrete would be hastened; and it would be 
 further accelerated if the temperature of its surroundings 
 was likewise increased by steaming. Moist heat, as 
 already pointed out, hastens the setting of concrete. The 
 surface of the concrete so laid may be further protected by 
 covering it with non-conductors of heat, such as straw, hay, 
 bags of sawdust, &c. 
 
 When concrete is well prepared, even in the ordinary 
 way, it is capable of supporting the action of the 
 London atmosphere successfully, as well as the action of 
 water. 
 
 The great point to be observed is to properly appor- 
 tion and to intimately mix the lime or cement into 
 the state of mortar, and then to present it to the 
 materials round which it is intended to crystallise in such 
 a manner that equality of setting may be secured. Well 
 selected blue lias lime, or grey stone lime, mixed with 
 pounded brick dust, will do well for the execution of con- 
 crete for the ordinary conditions of exposure to the weather. 
 But for exposure in sea-water the employment of Portland 
 cement is necessary. 
 
 Mr. E. Puscher, according to i\\t Journal of the Chemical 
 Society, March, 1883, recommends the following process 
 for rendering cement and lime less subject to atmospheric 
 influences : The cement materials should be allowed to 
 remain in a cold-water solution of one part ferrous sulphate 
 
in three parts of water for twenty- four hours, after which 
 they are dried in the air. The ferric oxide produced is 
 mechanically combined in the cement, and makes it denser, 
 harder, heavier, and weatherproof, filling up most of the pores, 
 .and giving it an ochre colour. Ornamental cement work is 
 brushed over with the solution four times and allowed to 
 dry. The cement work can be rendered extremely resisting 
 by warming and then coating with a hot mixture of equal 
 parts of paraffin wax in light petroleum. By treating twice 
 with a 5 per cent, soap solution, drying and polishing, the 
 surface is made efficient for oil painting. Chalk objects and 
 room walls treated in this manner will stand any amount of 
 washing. Light ochre colour can be obtained by adding 
 alum to the ferrous sulphate; and various shades of green 
 by painting with chrome-alum. 
 
 Hydraulic silica, gelatinous silica, and soluble silica 
 absorb lime gradually from lime water, the maximum 
 absorption varying in all cases, according to Mr. E. 
 Laudrin,* between 36 and 38 parts of lime for one equiva- 
 lent of silica. The resulting compound has approximately 
 the composition of 3 SiO 2 , 4 CaO. The combination is 
 most rapid in the case of soluble silica, but even in this 
 case the maximum absorption is not effected until after several 
 hours. Silica from hydrofluosilicic acid absorbs lime much 
 more slowly than any of the three previously mentioned 
 varieties. The maximum absorption after sixty-eight days, 
 in a series of experiments, was 24-2 parts of lime per 
 equivalent of silica. For this compound (3 SiCX, 4 CaO) 
 Mr. Laudrin proposes the name Pouzzo-Portland. When 
 mixtures of lime with different varieties of silica in the 
 proportion required to form Pouzzo-Portland are heated to 
 
 * Journal of the Chemical Society, August, 1883. 
 
156 
 
 bright redness in a gas-carbon crucible for a time, varying;, 
 with the nature of the silica, the fused, but non-vitrified 
 mass yields an artificial Pouzzo-Portland which generally 
 splits up and falls to powder as it cools. It is completely 
 soluble in hydrochloric acid, and when moistened with the 
 smallest possible quantity of water, and immersed under 
 water, it sets in from fifteen to sixteen hours, acquiring a 
 hardness, which, however, is scarcely equal to that of Spanish 
 white. If, however, the water is charged with carbonic acid, 
 the cement after some hours acquires a hardness equal to that 
 of the hardest stone. This fact tends to show that the 
 absorption of carbonic anhydride is an important factor in the 
 setting of hydraulic cement. 
 
 Specification for Concrete. There are numerous specifica- 
 tions for concrete, and it is impossible to lay down any 
 definite hard-and-fast rule, as the mixture of the ingredients 
 will necessarily vary according to the requirements of the : 
 case and the experience of the engineer. The following 
 are given as indications only which may be useful to the 
 reader. 
 
 Without specifying the use to which it is to be put, 
 Newbigging prescribes i part of lime, 4 parts gravel, and 
 2 parts sand.* 
 
 On the other hand, Mr. Boulnois gives the follow- 
 ing f: 
 
 " Street Foundations Heavy Traffic. 
 
 Portland cement i part 
 
 Sharp river sand 2 
 
 River ballast or broken stones ... 4 ,, 
 
 * Newbigging's " Handbook for Gas Engineers." 
 f " The Municipal and Sanitary Engineer's Handbook." Boul- 
 nois. Page 100, &c. 
 
'57 
 *' Liverpool Street Foundations. 
 
 Portland cement i part. 
 
 Gravel 5 to 6 parts. 
 
 Broken stone 7 to 8 
 
 The gravel and cement are thoroughly mixed dry, and only 
 enough water then allowed to flow on it to make the material 
 damp enough after it is incorporated to retain its form when 
 a portion is taken in the hand and squeezed.* 
 
 " The ground having been excavated, thoroughly consoli- 
 dated, and properly graded to the requisite shape, a layer 
 of broken stone, or other material, is spread evenly over the 
 surface and thoroughly wetted from the rose of a watering 
 <:an. A stratum of concrete, mixed as above, is spread 
 over this, and a second layer of stone added. The stone is 
 then beaten in with a heavy flat beater. Other layers of 
 mortar and stones are added and thoroughly beaten in until 
 the required thickness is obtained, the final layer of cement 
 concrete being smoothed off to an even and uniform sur- 
 face. After this concrete, or beton foundation has been 
 allowed to set ten days, the paving is commenced." 
 
 For footpaths, Mr. Boulnois recommends a concrete 
 made with i part Portland cement, 2 parts clean coarse 
 gravel, passed through lin. mesh, and 2 parts of clean 
 sharp sand. A finishing coat may be composed of i part 
 Portland cement, 2 parts granite chippings. 
 
 * This, it will be seen, is the Chinsse method. 
 
CHAPTER XIII. 
 
 ARTIFICIAL STONE. 
 
 History. To what era the introduction of artificial stone 
 belongs, it is not within the scope of this chapter to discuss. 
 Its antiquity, however, has been established by the archaeo- 
 logical researches carried out in Egypt, which have clearly 
 shown that at least one of the Pyramids was constructed of 
 artificial stone, /.&, concrete blocks made of small round 
 stones, broken stone, and lime. 
 
 In many of the old remains, still existing in England 
 from the periods of Roman occupation, can be distinctly 
 traced blocks of various shapes and sizes made from a con- 
 crete in which flint played a prominent part. 
 
 Coming to more recent productions, at the end of the 
 last century (eighteenth), a French architect carefully 
 worked out a process for making blocks of stone with a 
 special matrix, i.e. cementation principle, of cement which, 
 he made with lime, clay and charcoal. 
 
 Later, in England, just before the Victorian era, a Mr.. 
 Ranger took out several patents for the manufacture of a 
 concrete which was used in the wails of the Woolwich 
 Dockyard Wharf. Following this we have Coignet's " beton 
 agglomere," a composition of sand and hydraulic lime or 
 cement, and sometimes both. This was largely used in 
 France, but met with little favour in this country. Besides 
 sewers and numerous smaller works, blocks of it were used 
 for the Suez Canal, and a church at Yesinet, near Paris,, 
 having a steeple i35ft. high, was erected of it. 
 
 Ransomes' " Siliceous Stone " was perhaps the first 
 
'59 
 
 artificial stone which was prepared on a scientific basis. 
 Washed sand (i bushel) and a solution of silicate of soda,, 
 specific gravity 1-7 (i gallon), were mixed in a pug-mill 
 until the whole w r as thoroughly incorporated and w r as of the 
 consistency of putty, when it was moulded into blocks or 
 slabs, as required, and drenched with a solution of chloride 
 of calcium, then placed in tanks containing the same solu. 
 tion and heated to boiling. The resulting stone, the hard- 
 ness of which depended on the formation of calcium 
 silicate as a matrix, met with considerable success ; but 
 there was connected with it one unfortunate difficulty. It 
 was necessary for it to be prepared with extreme care a 
 care which only skilled labour was capable of and, as a 
 result, the process was found to be more or less com- 
 mercially impracticable and fell through. 
 
 Buckwell a few years later introduced a material which he 
 called " Granitic-Breccia " stone, which was composed of 
 Portland cement and oolitic or magnesium limestone, and 
 made by coarsely crushing the limestone, incorporating it 
 with the cement with the least possible quantity of water, 
 and ramming it into moulds. Owing to the retirement of 
 Buckwell, however, the operations fell through, but a stone 
 somewhat on the same principle is still made in consider- 
 able quantity, mention of which will be found later. 
 
 These two stones, Ransomes' and Buckwell's, have been 
 mentioned at some length, as they are practically the fore- 
 runners of all the modern artificial stones. Of these pro- 
 bably the best known is 
 
 The Victoria Stone. This stone was introduced by 
 Mr. Highton many years ago, but, handicapped at that 
 time by the variable quality of Portland cement, due to its 
 unscientific manufacture, and owing to experimental errors 
 with unsatisfactory materials, it did not at once meet with 
 
i6o 
 
 the success the material really deserved. About this time, 
 however, came a revolution in the manufacture of cement. 
 The Germans, studying the manufacture of Portland cement 
 from a scientific point of view, succeeded in making a 
 cement of practically constant composition and value, and 
 thus temporarily almost ousted their English rivals from the 
 market. As a result of this, the English makers, finding 
 their trade steadily declining, called in the best scientific 
 talent to assist them, and by the aid of combined research 
 succeeded in producing a cement which has practically no 
 equal. 
 
 With the use of a cement of practically constant compo- 
 sition, and a granite of exceptional hardness, the stone 
 rapidly rose in favour, and for over twenty-five years has 
 now held a prominent position. 
 
 This stone, which for many years was used almost 
 entirely for the making of paving stones, is a good example 
 of what can be done in concrete by the careful treatment of 
 suitable materials. 
 
 It is manufactured from a granite obtained from Groby, 
 in Leicestershire, the composition of which is : 
 
 Silica (insoluble) 65-26 
 
 Silica (soluble) 0-55 
 
 Alumina 13-06 
 
 Lime 4-55 
 
 Magnesia i-oi 
 
 Oxide of Iron 9-81 
 
 Soda 2*34 
 
 Potash 2-85 
 
 Carbonic Acid 0-03 
 
 Water, &c 0-54 
 
 The hardness of this granite is even greater than Guern- 
 sey granite, its crushing strain being about 20,750 Ib. per 
 
square inch. The granite is first finely crushed, and then 
 washed as free as possible from alkalies and alkaline salts,"" 
 and mixed in suitable proportions with Portland cement 
 generally about 3 parts of granite to i part cement in 
 the dry state by machinery, water is then carefully added, 
 care being taken to prevent elutrition, and whilst still plastic 
 it is placed in moulds, where, by shaking, pugging and the 
 judicious use of a trowel, a solid block is obtained free 
 from air-bells, and in such a state that the process of indu- 
 ration which it next undergoes will have the maximum 
 effect. When the concrete blocks thus formed have suffi- 
 ciently hardened, the moulds or frames, which are made of 
 wood lined with metal and fastened at the ends, are 
 removed and the blocks treated with a solution of alkaline 
 silicates in a bath for varying periods, generally about 
 fourteen days, after which they are exposed to the atmo- 
 sphere for some time tQ attain their maximum hardness a 
 hardness, of course, due to the combination of the silicate 
 with the free lime of the cement. 
 
 The stone thus produced has a chemical composition 
 of: 
 
 Silica 50 35 
 
 Alumina 1 1 87 
 
 Oxide of Iron 7 '33 
 
 Lime J 8'33 
 
 Magnesia 2 "03 
 
 Potash 1-78 
 
 Soda 3-81 
 
 Carbonic Acid 1-80 
 
 Water, &c 2-70 
 
 * Too much stress cannot be laid on the freedom of all stones 
 from alkaline salts, especially the sulphates of the alkalies. The 
 crystallisation and solution of these salts due to atmospheric ccn- 
 
 M 
 
l62 
 
 a crushing strain of about 8300 Ib. per square inch, about 
 30 per cent, more than Peterhead granite, a breaking strain 
 of about 1300 Ib. per square inch, and a porosity of 1*3 
 per cent, on a 24 hours' absorption test, being one-tenth 
 that of Portland stone, one-eighth that of Mansfield stone, 
 and practically equal to that of marble. The results 
 obtained by mechanical tests such as these give a very 
 good idea of the quality of a stone, still a test of actual 
 wear is worth recording. 
 
 In Islington, outside the Agricultural Hall, where the 
 traffic probably is nearly as heavy as any place in the 
 metropolis, this stone, after being laid for thirteen years, 
 had to be removed owing to alterations to the footway, &c., 
 but finding that it was in such good condition the local 
 authorities decided, on completion of the alterations, to 
 re-lay it on the reverse side, so that the new additions and 
 the old should be comparable. This was done some eight 
 years ago and the stone is in as good condition as ever. 
 
 As has been mentioned before, this stone was for some 
 time used only for paving, steps, window-sills, &c. ; of late, 
 however, considerable attention has been paid to the pro- 
 duction of a building material which should rival all ordi- 
 nary natural stones. This is made in identically the same 
 manner as the stone for paving, except that the size of the 
 crushed granite is slightly varied to suit the circumstances, 
 and that it is, if required, coloured by the addition of either 
 the yellow or red oxide of iron, to imitate the yellow or red 
 Mansfield stone respective y. 
 
 This introduction has met with a marked success a 
 success which is probably due as much to the way the 
 
 diticns cannot fail to finally cause disintegration of any material 
 containing them to any degree. 
 
material is handled as to the material ^itself. Lts extreme 
 hardness and closeness of grain renders it a most suitable 
 material for carving, the sharpness of the lines and contour 
 being all that could be wished. For this purpose the 
 blocks are cast to the approximate shape required,, the 
 Avhole of the carving being done with the chisel, as in 
 ordinary stone. 
 
 One important fact in connection with this stone is the 
 little effect weather and the atmosphere, even of manufac- 
 turing towns has on it. None, perhaps, have recognised 
 the defects of natural stone more thoroughly than those 
 who have been responsible for architectural work, or had 
 to build, near the sea. The] deplorable spectacle of clock- 
 towers, memorials, and hotel fronts pitted and weather- 
 -eaten is familiar to all a condition produced chiefly by 
 the crystallisation of the salt within the pores of the stone, 
 and the consequent flaking, &c. The low porosity and hard 
 ness of this patent stone enables it to resist the action of even 
 the sea air a fact which is now beginning to be recognised 
 and taken advantage of, the substitution of artificial for 
 natural stones being noticeable all along our coasts. 
 Amongst others the Terriss Memorial at Eastbourne, 
 .and Morley House Convalescent Home at St. Mar- 
 garet's Bay, the entrance to which is an excellent 
 -example of what can be done in architectural reproduction, 
 is built of this stone, and has fully justified the opinion of 
 those who advocate the advantages of artificial over natural 
 stone. 
 
 One cannot but look favourably on a stone such as this, 
 which, less costly than even the soft Bath stones from which 
 all detail is effaced in a few years, free from the too 
 frequent flaws in natural stone, flaws which unfortunately 
 only show up after some little wear, ready of production 
 
 M 2 
 
164 
 
 and constant of composition and durability, is made on a 
 sound scientific basis. 
 
 Ward's Stone. Under this name are two distinct stones. 
 The first, which calls for but slight mention, is made from 
 crushed granite and Portland cement, somewhat on the 
 principle of the " Victoria Stone," except that no process 
 of induration is used, the desired effect being claimed to be 
 obtained by using a special cement which prevents air 
 cracks, &c., in drying. This stone is largely used for 
 paving, stable flooring, &c. 
 
 The other stone has been mentioned before in the open 
 ing part of this chapter. It is practically a modernised 
 modification of BuckwelFs " Granitic Breccia " stone. 
 
 For the manufacture of this material various limestones,, 
 chiefly of the oolitic variety, are coarsely crushed, mixed 
 with a special Portland cement, and cast in moulds or 
 frames, according to the block required. When thoroughly 
 set and hardened, the surface is rubbed down to whatever 
 design is required and finally polished. Slabs or mouldings 
 in any form, or anything which can be cast to the approxi- 
 mate size and form required are suited to it, but it is not 
 suitable for carving on account of the varying hardness of 
 the different particles of limestone, &c., set in the 
 matrix. 
 
 In its finished state the stone bears a marked resemblance 
 to, and, in fact, might be considered as an imitation Belgian 
 stone, it takes a remarkably good polish, considering the 
 nature of the material, yet is not and does not become 
 slippery. Owing to its decorative properties, combined 
 with its durability, it has been largely used for ornamental 
 staircases, the steps, &c., to the chancel of the new Roman 
 Catholic Cathedral at Westminster being amongst the 
 many uses to which it has recently been put. 
 
Granolithic Stone Stuart's. This is a noteworthy 
 example of the granite and cement-matrix stone. Particular 
 attention has-been paid to the production of a stone capable 
 of resisting the continuous action of water, and to its 
 adaptation to. fireproof work two extremes which speak .well 
 for the material. Probably no more severe test could be 
 applied than intermittent submersion, a test which only 
 picked natural stone will withstand. The successful con- 
 struction of docks, such as the Glasgow graving dock, and , 
 water terraces is, perhaps, the best answer that could be 
 given to critics who seek to disparage artificial stone. 
 
 " Non-slip" Stone. For many years hard York stone was 
 regarded as the standard material for paving, &c. ; but, 
 owing to the laminations which are inseparable from it, 
 together with its irregular size and thickness and variable 
 durability, it came to be looked on as far from an ideal 
 standard. The constant trbuble caused by this material to 
 some extent prepared the way for the introduction of an 
 artificial stone which could be obtained in regular sizes, 
 even thickness, and square edges throughout the full 
 thickness. 
 
 Recognising that it was impossible to obtain a surface 
 which did not become more or less polished, and therefore 
 not have the foothold of York stone, when granite was 
 used as the base of the artificial stone, one of the large York 
 stone companies took out a patent for the manufacture of 
 an artificial stone, using York stone chippings as its 
 base. 
 
 This material, which has been named " non-slip '' stone, 
 is made by crushing hard York stone chippings, and 
 intimately mixing them with a suitable cement, casting the 
 mixture into moulds and submitting it to hydraulic pressure. 
 By this means a stone is produced having a surface practi- 
 
i66 
 
 cally the same as York stone, and a greatly increased 
 durability, besides the advantages already mentioned. 
 
 Moreau Marble. In nearly all the artificial stones that 
 have been introduced Portland cement forms an essential 
 ingredient ; practically, therefore, they are concretes. 
 
 In addition to these concretes, artificial stones have from 
 time to time been introduced, with more or less success,, 
 having for their basis the changing of unsuitable natural 
 materials into hardened materials closely resembling natural 
 stones. 
 
 Probably the most noteworthy of these is one to which 
 the name " Moreau Marble " has been given, the formation 
 of which, although cementation undoubtedly plays a part in 
 it, does not depend on Portland cement for its cohesion. 
 
 This stone the name marble is somewhat of a misnomer,, 
 although the finished article bears the closest resemblance 
 to it can be produced from practically any amorphous- 
 variety of limestone, provided it is sufficiently soft, varieties- 
 which are worthless for building purposes. 
 
 The limestones which, however, are usually used in the 
 manufacture of this " marble " are of two varieties, both of 
 which have great uniformity of structure and evenness of 
 surface. One variety is obtained from the neighbourhood 
 of Poictiers, and the other, a fossiliferous limestone, from, 
 Angouleme, in both of which places huge deposits occur. 
 
 The process of formation is simple, the limestone being; 
 first cut to the approximate size, and then planed down by 
 means of a steel cutter to the moulding or surface required. 
 
 The requisite shape having thus been obtained, the stone 
 is then veined by a method somewhat similar to that used 
 in chromo-lithography, the veins which it is desired to- 
 reproduce in the stone being first of all rendered impervious 
 by the use of a suitable varnish, which, soaking into the 
 
i6 7 
 
 stone, prevents that part from absorbing the colour which is 
 next applied. 
 
 In order jto produce more natural veining than it would 
 be possible to obtain by hand, a solution of ' gum Thus" in 
 turpentine, with the addition of a little insoluble colouring 
 matter, is floated on the surface of water and spread by a 
 spray of soapy water. The face of the stone is then placed 
 on the surface of the water, and thus picks up the streaks 
 and patches which afterwards become the veins, &c., of the 
 finished stone. 
 
 The stone is next coloured by immersing it in one or 
 more solutions of metallic salts for different periods accord- 
 ing to the shades required, thus producing a coloured 
 stone by the same material with which the natural marble is 
 coloured, and finally, if necessary, the colour is fixed by hot 
 water. 
 
 Having thus obtained tne coloured and veined stone, it 
 now only remains to harden the material, which is done by 
 immersion in a solution of sulphate of zinc for a period 
 generally about 24 hours. 
 
 It is somewhat difficult to understand why sulphate of 
 zinc should have such an effect ; certain it is that practically 
 the whole of the zinc introduced is converted to carbonate, 
 yielding an equivalent quantity of calcium sulphate, and 
 probably the zinc carbonate thus formed acts as a stopping, 
 filling in the interstices in the loose molecular structure of 
 the limestone, whilst the resulting calcium sulphate crystal- 
 lising out slowly in the structure sets up a state of tension 
 in the stone, which still further tends to harden it. How- 
 ever, though the action is uncertain, the result is highly 
 satisfactory. The stone, after its hardening bath, is dried 
 first in a chamber at about 122 deg. Fah. (50 deg. Cent.), 
 and finally in a chamber at about 212 deg. Fah. (ico deg. 
 
i6S 
 
 Cent.), when it is ready to receive its polish in the same way 
 as ordinary marble. 
 
 When it is required to imitate Rouge Royal marble, or to 
 produce a "marble" which it is known will withstand con- 
 tinued heating, the stone is coloured by salts of iron, and 
 treated in the same way, except that after drying it is 
 exposed to a heat of about 800 deg. Fah. (446 deg. Cent.) 
 in an oven. 
 
 The stone thus produced, as mentioned before, is ex- 
 ceedingly like in appearance to an ordinary marble, and 
 takes a polish quite equal to it. 
 
 Too much stress cannot be laid on this point, as the 
 higher the initial polish on any carbonaceous stone the 
 better able it is to withstand the attack of the sulphurous 
 and sulphuric acids present in the atmosphere of every 
 town. 
 
 It has a hardness closely approximating to marble, and a 
 crushing strain of about 4100 Ib. per square inch, the 
 crushing strains of marbles varying from 224olb. to about 
 6000 Ib. per square inch. 
 
 In carving, sharpness of detail, and effect, it certainly is 
 equal to marble. In only one instance, probably, does it 
 fall short of marble ; the initial cream colour of the lime- 
 stone proves an insurmountable difficulty, and renders it 
 impossible to reproduce the white marbles used for sculp- 
 ture, &:c. 
 
 Over its compeer it certainly has several advantages. 
 Slabs and columns that in the foreground of the illustra- 
 tion is in one piece which would be prohibitive in marble, 
 could readily be made in this stone. 
 
 It is free from the objectionable " stopping " which, fcr 
 economic reasons, is so prevalent in many marbles ; it is 
 much less costly, and can be produced in a comparatively 
 
169 
 
 short space of time. In an ordinary way the stone can be 
 prepared and the process completed in twenty-one days ; 
 that erected in the entrance hall of the Balham Theatre 
 being actually completed and fixed in sixteen days. 
 
 Having regard to the possibilities of colour, design, &c., 
 it is perhaps somewhat to be regretted that so much atten- 
 tion is paid to making close imitations of marble. With 
 such illimitable advantages of schemes of colour, easiness 
 of working, and size, it would certainly seem better for the 
 material to be treated more on its own merits. 
 
 It certainly should not be considered an imitation marble ; 
 it does not possess marble's essential feature, crystalographic 
 form. It is a stone standing by itself, possessing all 
 marble's advantages and more, but it commences the 
 process as a stone and ends as a stone. 
 
 Terra - cotta. Any account of artificial stone would 
 scarcely be complete without mention of what is probably 
 the most enduring of building materials, namely, terra- 
 cotta, although it cannot by any means be considered as an 
 artificial stone any more than a brick can. Yet on account 
 of its process of manufacture its inclusion needs no 
 apology. 
 
 Dating from many centuries B.C. its peculiar applicability 
 "has only been fully recognised during the latter end of the 
 nineteenth century, mainly owing to misconceptions which 
 arose as to its mode of architectural treatment. 
 
 Spread over the surface of the world are huge deposits 
 of clay more or less suitable for the manufacture of terra- 
 cotta, amongst those particularly adapted for the purpose 
 being the North Devon and Dorsetshire clays, which con- 
 tain in small quantities the alkalies which are necessary to 
 aid the vitrification of the surface when the material is fired, 
 and having for their composition ingredients which render 
 
170 
 
 them suitable for working, practically without admixture. 
 These clays have the following composition *: 
 
 North Devon. Dorsetshire. 
 
 Alumina .. .. 29-28 .. .. 32-11 
 
 Silica 52-06 .. .. 48-99 
 
 Lime 0-43 .. .. 0-43 
 
 Magnesia .. .. 0-02 .. .. 0-22 
 
 Oxide of Iron .. 2-37 .. .. 2-34 
 
 Potash 2 - 29 . . . . 2-31 
 
 Soda 2-56 .. .. 2-33 
 
 Water of Combina- 
 tion 1 0-27 . . . . 9-60 
 
 The clays obtained from some of the coal measures are 
 also found, when properly used, to be most suitable. 
 
 The principal makers of terra-cotta have found, however,, 
 that they obtain more satisfactory results by a judicious- 
 mixture of clays, &c., a process which, though more expen- 
 sive than using the natural clay, they consider amply 
 justified by the improved result. What probably gives the 
 manufacturer most trouble in the production of this material 
 is the shrinkage which takes place in drying and firing. In* 
 order to reduce the amount of shrinkage and to aid in the 
 drying, it has been found desirable to add refractory 
 materials, such as previously hard-burnt clay, to the raw clay 
 and ground with it. The manufacture may briefly be 
 summed up as follows : 
 
 To clay which has been carefully mixed and brought up- 
 to a standard of known shrinkage by the addition of re- 
 fractory materials is added a vitrifying agent, when, after 
 again carefully mixing with water to the required degree of 
 plasticity, the prepared clay is ready to be pressed into the 
 moulds which give it the desired form, and then dried 
 by hot air, and fired in a suitable kiln. 
 
 * Analyses by Weston. 
 
In the use of terra-cotta several points must be kept in 
 mind. The surface of this material being slightly vitrified 
 is the most resistant, and therefore care must be taken to- 
 preserve it intact, i.e., free Irom chips ; it should be suffi- 
 ciently porous at the joints to make a perfect bond with 
 the mortar ; great care should be taken that all exposed 
 joints are made good with cement mortar an absolute 
 necessity with this material. Large columns and, therefore, 
 classic architecture, are unsuited to the material ; designs 
 should not call for larger blocks than 2 cubic feet, and 
 should generally not exceed ift. cube. Any detail can 
 be reproduced, but nothing produces such effect as that 
 which shows the hand of the modeller of the natural 
 material clay. One of the earliest examples of the use 
 of the material for a large public building in modern times 
 may be seen in the Natural History Museum at South 
 Kensington. The architectiTral outlines as seen from a 
 distance are particularly striking, whilst on closer inspection 
 one is charmed by the wealth of detail 
 
 For many years terra-cotta was kept back and nearly ruined 
 by the specifications of architects, who, either from lack of 
 artistic taste, or because their specifications had always been 
 drawn in this way for stone, &c., demanded uniformity of 
 colour. 
 
 No greater blow could have been aimed at this industry 
 than such a demand. The struggle for uniformity of colour 
 tended towards under-firing the material, with consequent 
 weakness and perishability, at the same time robbing it of 
 all its natural qualities of play of colour and evidence of 
 its presence in the fire. In the same way architects, with- 
 out studying the nature of the material they had to deal 
 with, treated it as if it were an ordinary stone, and pre- 
 scribed for it all sorts of impossible purposes, then, finding 
 
172 
 
 it could not be made to conform to their ideas, proceeded 
 to damn it. 
 
 With the increasing recognition of its value, however, a 
 number of architects have, by making a special study of 
 the subject, and by availing themselves of the help of ex- 
 perts in the employ of the larger manufacturers, succeeded 
 in demonstrating many legitimate treatments of archi- 
 tectural subjects in this material, and even in some cases 
 have produced beautiful buildings with characteristics which 
 could hardly have been obtained in any other material. 
 The delicate divergencies of colour have been expanded, 
 and schemes of architecture suited to its peculiarities 
 developed, with the result that buildings have grown up 
 around us possessing artistic individualism and affording a 
 pleasant relief to the gloomy repetition so prevalent in our 
 great cities, for which reason, if for no other, the develop- 
 ment of terra-cotta is to be welcomed. 
 
 The foregoing summary includes the most important 
 artificial stones now in use. Besides these, from time to 
 time a number of different stones have been introduced, 
 having in common with most of those described a cement 
 matrix with a varying base, such as crushed coke, slag, c. 
 During the last few years several patents have been taken 
 out for the production of a stone on an entirely different 
 basis to any of the above, being made from silica, in the 
 form of fine sand, and lime only. Although varying in 
 detail, the principle is the same, i.e., the combination of the 
 lime with the silica by superheated steam under several 
 atmospheres pressure. By this means blocks consisting of 
 silicate of lime, silica and lime, have been made possessing 
 to all appearances the characteristics of natural stone, but 
 in reality having many points in favour which natural 
 stones never have or can have. 
 

 173 
 
 The stone produced by one of the most recent patents 
 has, indeed, the peculiar defect that unless worked within a 
 few days it becomes so hard that ordinary masons reject it 
 as being useless to them, they being unable to deal with it 
 with their ordinary tools, When, however, the details con- 
 nected with its manufacture have been completed it will 
 undoubtedly take a high standing as a building stone. 
 
 To anyone who has studied the subject of artificial stone 
 it becomes a matter of wonder why, possessing the many 
 advantages it does, it is not more generally used. I believe 
 the real answer to this may be found in the general conser- 
 vatism of the architectural profession ; trained as they are to 
 architectural styles and laws handed down from generation 
 to generation, they naturally regard materials some\yhat in 
 the same light, and so, in copying classic monuments of 
 architecture, they try to obtain stone as closely resembling 
 the original as possible. 
 
 Their one reply to such a question is that they have 
 always used natural stone, and they prefer it. No reason, 
 no argument ; they have to prescribe, and they prescribe 
 what their grandfathers did before them. 
 
 Doubtless, however, some prejudice has arisen against 
 artificial stone owing to the action of unscrupulous mer- 
 chants, who have put on the market worthless rubbish, 
 generally with a name somewhat similar to some well-known 
 make. A very simple rule, however, eliminates any such 
 possibility as this, and is therefore worth mentioning : 
 insist on seeing how the material is made that you are 
 going to use, and if it is a genuine stone every facility will 
 be granted you ; if a fraud you will be put off with garbled 
 information. 
 
 Year by year artificial stone is gaining in popularity, and 
 studying its possibilities, one can scarcely doubt its future. 
 
CHAPTER XIV 
 
 ASPHALT. 
 
 BY the kind permission of Mr. Clifford Richardson, of 
 Long Island City, New York, I am enabled to present to 
 my readers the following most interesting and valuable 
 account of the nature and origin of asphalt. These re- 
 searches are contained in a seiicu ui * Contributions to the 
 Chemistry of the Natural Hydro-carbons and their Deriva- 
 tives," from the laboratory of the Barber Asphalt Paving 
 Company, and have been re-printed, with corrections and 
 additions, from an article in the Journal of the Society of 
 Chemical Industry for January, 1898. 
 
 Form of the Lake Deposit 
 
 All that was known in 1891 of the Pitch Lake deposit was 
 purely superficial, and showed merely that it covered an 
 area of about 114 acres; that the surface was in constant 
 motion, as was proved by the appearance on its surface of 
 stumps of trees brought up from below, which, after being 
 carried to a certain height, would topple over, to be again 
 engulfed and disappear ; that there were several groups of 
 trees or islands sustained on the pitch ; that there was a 
 pool of soft pitch of ordinary temperature at the centre 
 where gas was evolved ; and that wherever a hole was dug 
 in the pitch, whether deep or shallow, it filled up, and the 
 surface resumed its level after a short time. Nothing was 
 known as to the size or depth of the deposit, the shape of 
 the enclosing walls, or of the amount of new material 
 which flowed into the lake each year. 
 
In 1893 and 1894, however, when the Trinidad Asphalt 
 Company began to establish a pier and cableway for con- 
 veying the crude pitch from the lake to vessels for shipment, 
 a series of borings was made upon the lake by Mr. P. W. 
 Henry, now general manager of the Barber Asphalt Paving 
 Company, as well as on the land between the lake and the 
 .gulf, and in the gulf to the north and west of the lake ; 
 and a line of levels was run from the shore to the lake, and 
 across it through the centre in several directions, with 
 stations at intervals of iooft, all secured by bench marks on 
 firm ground. 
 
 The boring at the centre of the lake was carried to a 
 depth of i35ft., the entire distance being through pitch, 
 which, as far as ocular evidence goes, has the same 
 character as that at the surface. It was impossible to carry 
 the boring deeper, as the movement of the pitch had so 
 inclined the tube ift. in 6ft. which formed the lining, 
 that it had to be abandoned. It then gradually toppled 
 over and was engulfed. Nothing has been seen of it since. 
 The result was sufficient, however, to show the great depth 
 of the crater and the uniformity of the pitch. The depth 
 attained was within a few feet not more than three and a 
 half of sea level, and yet we do not know how much 
 deeper the pitch may extend. The borings on the north 
 side of the lake, about loooft. from the centre, and iooft. 
 from the edge, was in pitch of the usual character for 75ft., 
 showing a very steep slope to the sides of the crater. At 
 Soft, a layer of fine white sand was met for a few feet, 
 and then asphalt was again encountered. At 9oft. sand 
 mixed with asphalt was struck, and this continued to a 
 depth of 1 5 oft. 
 
 Further borings, made at some distance from the lake, 
 gave results near the surface which were similar to those 
 
i 7 6 
 
 found at the deeper levels at the edge of the lake. Sand, 
 mixed with asphalt here and there, was the common material, 
 while at a depth of Soft, on the southern side of the lake, 
 and about Soft, south of the road, and between i2ooft. and 
 i3ooft. from the centre of the lake, a very hard asphaltic 
 sandstone was found. 
 
 All the evidence thus goes to show that the sides of the 
 crater are of sand or sandstone, more or less impregnated 
 with bitumen, the sandstone being no doubt the rock of the 
 hillside toward the south, against which the crater has been 
 built up. 
 
 From the borings it was thus learned for the first time 
 how enormous the deposit was, and the idea that the mound 
 was really a crater seemed to be confirmed. It is, never- 
 theless, hard to realise that there is at this point, 1380:. 
 above the sea, a bowl-like depression over 23ooft. across, 
 and over 135 ft. deep, reaching below the sea level, and 
 filled with a uniform mass of pitch, which must amount to 
 over 9,000,000 tons. Nothing less remarkable are the con- 
 clusions reached from the results of the lines of levels run 
 across the lake. 
 
 In February, 1893, four lines of levels crossing at the 
 centre were run across the lake and secured by plugs on 
 the shore. Bench marks of concrete were also put in on 
 hard clay some distance back from the lake, which in 1894 
 were found not to have changed their level. The results of 
 these levels of 1893 showed that the centre of the lake was 
 about a foot higher than that portion a thousand feet out- 
 ward toward the edge, and that from the latter point to the 
 edge was a rise of 6in. The elevation of the centre above 
 sea level was i38*5ft.; of the station, loooft. N., 27 deg. W., 
 i37'5ft; of the edge, nooft [from the centre, 138'oft.; 
 while the top of the crater wall itself, looft. further on, was 
 
'77 
 
 140'ift, and at the diametrically opposite side of the lake, 
 141 '4ft. There is considerable irregularity in the height of 
 the crater wall, due, no doubt, as suggested by Peckham, to- 
 its breaking down in part under the pressure of the pitch. 
 The highest part, as shown in my 1892 report, is to- the 
 south, with an elevation of 141 ^ft. above sea level, and 
 the lowest toward the west and toward the north-east r 
 where it is not less than 137 '8ft. or 13811. above sea level,, 
 the highest level probably being that of the original rim, 
 as confirmed later. 
 
 The surface of the lake, it is very evident from even a 
 casual examination, is lower to-day than some years ago, 
 and the deposit now seems to occupy a shallow depres- 
 sion. This fall is due to the removal of the crude pitch 
 for shipment, which has reached over a million tons. 
 From the difference in level of the surface of the lake 
 between any two years, and*from the amount of pitch 
 removed, it is easy to calculate how many tons of it corre- 
 spond to a fall of an inch or a foot. In a similar way, from 
 the area of the surface and the density of the pitch, it .is 
 possible to calculate on another basis how much pitch 
 should correspond to an inch or a foot in depth. From 
 the latter figures it appears that the lake should have fallen 
 many more feet than it has in the last thirty years, and from 
 a comparison of the results of the first and second calcula- 
 tions, extending over three periods and three sets of levels,, 
 there is no doubt that there must be an influx of fresh 
 pitch at the soft spot, which, in the periods between 1893. 
 to 1896, amounted to an average of from 20,000 to 18,000- 
 tons per year. There is, therefore, no doubt that there is a 
 vast influx of pitch into the lake at the present time,, 
 amounting to at least 18,000 long tons per year, and adding 
 to a supply which must, from the area and depth of the 
 
i 7 8 
 
 crater, reach 9,000,000 tons. The enormous size and 
 activity of this deposit are equally striking. 
 
 Movement of the Lake Surface. 
 
 The constant movement of the surface of the lake, as 
 shown by the course which sticks and logs take, which rise 
 in the pitch, was well brought out in the work of running 
 the levels which have been mentioned. Stakes driven for 
 stations in levelling in a right line across the lake were, near 
 the more actively moving central portion, much out of line 
 in twenty-four hours, and within three weeks those at 
 intervals of looft. for 6ooft. from the centre had moved as 
 follows : 
 
 Centre o . . 20 -6ft. to right . . 12-7^. ahead on line 
 i I3-5 >. i*i 
 
 ,, 2 6-5 ,, II'O ,, ,, 
 
 ,- 3 i'7 .1 -o 
 
 ,. 4 ' 2 2-4 
 
 ,, 5 0-9 3-8 
 
 . 6 3-2 1-4 
 
 On one of the lines there was an island or mass of floating 
 vegetation which was marked by a hub in 1893. In 1894 
 this island, 6ooft. from the centre of the lake, had moved 
 5 '5ft. to the left of the line, and 23ft. in the direction of the 
 line toward the edge of the lake. The positions of the 
 islands intersecting the lines of levels were also deter- 
 mined in 1896, and all of them found to have shifted their 
 position. 
 
 By these determinations, therefore, the activity of the 
 surface movement of the lake is definitely settled. 
 
 Proximate Composition of Trinidad Asphalt. 
 
 When in Trinidad, in 1891, Mr. Richardson made a 
 collection of specimens of the crude pitch, or asphalt, at 
 
intervals of 2ooft., on lines laid out on two diameters of the 
 lake. These were examined according to the methods in 
 use at that time, the water having been previously removed 
 by drying at 100 deg. Cent., and the solvents being applied 
 at ordinary temperatures. The results obtained showed 
 great uniformity in the composition of the crude asphalt, 
 but were not entirely satisfactory for several reasons. The 
 specimens were taken too near the surface of the lake; they 
 were liable to have suffered an alteration and loss of light 
 oils by being dried at too high a temperature; and the more 
 recent methods of analysis, with the use of light naphtha as 
 -a solvent, had not been elaborated. He found soon after 
 that the emulsified water in the pitch could be removed very 
 readily, without the aid of heat, by grinding the pitch to a 
 fine powder and exposing it to the air. The rapidity with 
 Avhich the water evaporates and the ease with which the 
 pitch dries is illustrated by the following determinations. A 
 piece of crude pitch was weighed, ground, and passed 
 through an 8o-mesh sieve, and then exposed in a thin layer 
 *o the air of the laboratory. The loss of water was as 
 follows : 
 
 Loss in Percent P "-Lrf entire 
 5 minutes 2*2 7-0 
 
 12 3'5 I2'I 
 
 20 ,, 6'5 22'4 
 
 30 9-5 32'7 
 
 1 hour 16-5 56-9 
 
 2 hours 20 -o 69-0 
 
 3 >. 21-0 72-4 
 
 and after regrinding 
 
 4 hours .. .. .. .. 22-5 77 -6 
 
 24 ,, 29-0 100-0 
 
 and in vacuo over H 2 SO 4 o'6 
 
 On exposure to air saturated with moisture, the powder of 
 
 N 2 
 
i So 
 
 crude pitch, after final drying in vacuo, gained in twelve 
 hours 2 '8 per cent, an amount of hygroscopic moisture 
 which is not large, and which is quite different from the 
 emulsified water originally present, which evaporates as soon 
 as the bituminous cells which enclose it are broken down. 
 In fact, the manner in which the water exists in Trinidad 
 asphalt is quite different from that in which it is found in 
 most other substances occurring in Nature, since it cannot 
 be removed by diffusion or osmosis, as in the case of drying 
 a lump of clay or a vegetable structure, but only by breaking 
 down the enclosing cell wall of bitumen. The novelty of 
 this method of drying the crude asphalt was acknowledged 
 by the Patent-office, and a patent granted therefor. The 
 peculiarity has since been noticed by other investigators. 
 Owing to the ease with which the water can be removed 
 from Trinidad Lake asphalt in this way, a means was afforded 
 of collecting the pitch and drying it at the lake, so that no 
 change could take place in transit. 
 
 A final series of samples was collected in this way by Mr. 
 P. W. Henry during his work at the lake in February, 1894. 
 In order also to obtain true representative samples of the 
 material forming the mass of the deposit, the surface samples 
 were taken about 2ft. below it, and others from the borings 
 which have been mentioned as extending as deep as i35ft. 
 at the centre and i5oft. on the edge. Mr. Richardson was 
 thus supplied with a set of samples which fairly represent 
 the true character of the crude pitch as it exists in the lake.. 
 These samples were examined as follows : 
 
 Method of Analysis. 
 
 Separate weighed portions of the material dried in vacuo 
 over sulphuric acid were extracted in beakers with successive 
 portions of hot carbon bisulphide and naphtha of 88 deg. B.P f> 
 
iSr 
 
 the decanted solvents being passed through a Gooch 
 crucible with heavy asbestos felt. The filtrate was allowed 
 to settle, in the case of the carbon bisulphide extract for 
 twenty-four hours, again decanted, and any fine sediment 
 which had passed the filter brought upon it. The losses 
 represented the bitumen soluble in the two solvents. The 
 mineral matter was determined by direct ignition, and the 
 organic matter not soluble by difference. This determina- 
 tion by difference is too large, owing to too great loss on 
 ignition in the determination of the mineral matter ; and 
 too small, owing to the calculation as bitumen of some 
 substance that is left in the bisulphide solution which is not 
 bitumen. The two errors nearly neutralise each other, 
 although in more recent analyses the carbon bisulphide 
 solution is finally burned and ignited, and the amount of 
 mineral matter found subtracted from the loss taken as the 
 amount of bitumen. 
 
 Great care is essential in these determinations, especially 
 that the solvents be perfectly dry. On this account 
 bisulphide is much more suitable for. use than oil of turpen- 
 tine or chloroform, as it is not nearly so hygroscopic, 
 although it is not quite as complete a solvent unless used 
 hot. 
 
 Examined in this way, the results given on page 182 were 
 obtained for the 1893 collection. 
 
 In these analyses of the surface samples an even more 
 striking uniformity in composition is found than in those 
 collected in 1891, owing to the care in their collection and 
 preparation, and to the method of analysis. The amount 
 of bitumen is somewhat higher, on account of the more 
 thorough extraction (to day, with still further improvements 
 in the mode of treatment, it would be even more so), but 
 .relatively Mr. Richardson found the same evidence of a 
 
182 
 
 fixed proportion of bitumen, mineral matter and organic 
 matter not bituminous. 
 
 In the samples from the boring at the centre of the lake^ 
 which extended to a depth of issft., and was still in? 
 asphalt, there is not as great uniformity, because of the 
 way in which it was necessary to collect them by washing. 
 the particles detached by the boring machine up through 
 
 Average Composition of Trinidad Lake Pitch in Circles. 
 
 Bitumen , 
 From centre. by 
 CS 2 . 
 
 Vlineral 
 matter. 
 
 Percentage 
 Organic, Soluble of total 
 not in bitumen 
 soluble. Naphtha. thus 
 soluble. 
 
 Percent. Percent. Percent. Percent. 
 
 Circle 2, 
 
 200ft. 
 
 55 
 
 02 . . 
 
 35 
 
 '4 1 
 
 .. 9 
 
 '57 
 
 3i 
 
 83 .. 
 
 57' 
 
 '8> 
 
 Circle 4, 
 
 4Ooft. 
 
 54 
 
 '99 
 
 35 
 
 40 
 
 9 
 
 61 
 
 3 1 
 
 '63 .. 
 
 57 
 
 '55- 
 
 Circle 6, 
 
 6ooft. 
 
 54 
 
 84 .. 
 
 35 
 
 '49 
 
 .. 9 
 
 67 
 
 3 1 ' 
 
 85 .. 
 
 58 
 
 26. 
 
 Circle 8, 
 
 800 ft. 
 
 54 
 
 66 .. 
 
 35 
 
 56 
 
 .. 9-78 3i 
 
 67 .. 
 
 57' 
 
 '97 
 
 Circle 10, 
 
 loooft. 
 
 54 
 
 78 .. 
 
 35 
 
 '44 
 
 .. 9 
 
 78 
 
 3 1 
 
 58 .. 
 
 57' 
 
 64 
 
 Circle 12, 
 
 1 1 oof t. 
 
 54 
 
 62 .. 
 
 35 
 
 '45 
 
 .. 9 
 
 93 
 
 3 1 ' 
 
 '77 
 
 57' 
 
 5 1 
 
 General average 
 
 54 
 
 92 .. 
 
 35 
 
 46 
 
 .. 9 
 
 72 
 
 .. 31 
 
 72 .. 
 
 57' 
 
 79 
 
 Circle 14, 
 
 I40oft. 
 
 53 
 
 86 .. 
 
 36 
 
 38 
 
 .. 9 
 
 76 
 
 .. 30' 
 
 52 
 
 56- 
 
 66. 
 
 Average Composition of Trinidad Lake Pitch from the 
 Boring. 
 
 54' 66 35'9o 9'44 3I-53 57' 6 7 
 
 the bore with a current of water, and catching the material, 
 in a pail, where x it was allowed to settle and the sample, 
 taken. Nevertheless the results are sufficiently close ta 
 show that the material at all depths is the same, and 
 when they are averaged the agreement between the com- 
 position at the surface and for an average depth is re- 
 markable. 
 
i*3 
 
 The average composition of the two lots of sample 
 from the surface and from the boring is as follows : 
 
 v Bitumen ~. . 
 
 Bitumen. Mineral. Organic, soluble in , . 
 
 Percent. 
 
 Surface .. 54-92 .. 35-46 .. 9-72 .. 31-72 .. 57-79 
 Boring .. 54-66 .. 35-90 .. 9-44 .. 31-53 .. 57-67 
 
 At the side of the lake the boring, about xooft. from the 
 edge of the crater, showed similar pitch to that on the 
 surface for 75ft., after which there was, as has been men- 
 tioned, a change, and the composition of the borings 
 proved to be that of sand and soil mixed with asphalt, 
 showing that the side of the crater had been reached. 
 
 The uniformity of the proximate composition of all the 
 pitch that exists in the lake and is newly forming there 
 leads to the belief also that the name Parianite, suggested 
 for it by Prof. Peckham, as "a mineral species, is entirely 
 justified. 
 
 Bitumen Soluble in Naphtha. 
 
 In Mr. Richardson's report of 1892 he showed that 
 between the refined products of that pitch from the 
 Trinidad Lake and that from deposits outside of the lake, 
 known as "land asphalt," a decided difference could be 
 detected in the relative amount of their total bitumen 
 soluble in 88 deg. naphtha. 
 
 Determinations, in the same way, of the amount soluble 
 in the lake samples, have proved of interest in connection 
 with the comparison of those from the surface and from 
 the borings. The average percentage of bitumen soluble 
 in naphtha, in both the surface samples and those from the 
 boring at all depths, is the same, or nearly so 5779 and 
 
i8 4 
 
 57-67 per cent. ; but on the surface itself the amount is 
 found to diminish somewhat toward the edge of the lake 
 and in the boring the average for the first 7oft. is 56-83 
 per cent., while for the second it is 58-50 per cent. The 
 effect of age is seen in these cases much in the same way 
 as, to a much greater extent, is found to be the case with 
 the land asphalt. 
 
 The Deposits of Land Asphalt. 
 
 Beyond the boundaries of the Trinidad pitch lake, and 
 between it and the sea, and even in the sea itself, are 
 found deposits of crude pitch very similar to that in the 
 lake, or in some stage of alteration which permits of their 
 recognition as being originally derived from such 
 material. 
 
 There is no doubt that a large area to the north and east 
 of the lake contains large quantities, and it has been an 
 open question as to how it got there. Manross describes 
 an overflow from the lake, and to this is undoubtedly due 
 the presence of pitch on the lands adjoining it and for 
 some distance toward the sea to the east, but a large 
 amount of the pitch must have been ejected independently 
 of the lake source of supply, but so long ago that it has 
 been buried with soil for years. There has been no over- 
 flow for years, nor any evolution of fresh pitch at any point 
 in the neighbourhood of the lake, except one or two small 
 cones, so that there is no activity to-day which will reveal 
 what happened in past centuries. The pitch has reached 
 the seashore in many instances, and even forms reefs 
 beyond it. It is spread out on the Point d'Or estate to 
 the east of the lake, evidently in a large sheet, which seems 
 more like an evolution of pitch independent of the lake 
 
supply than any other. This has been overgrown for some 
 time, and its age, which can only be determined by the 
 changes which analysis would show it to have undergone, is 
 doubtful. 
 
 Where the pitch comes in contact with salt water it is 
 hardened and does not rot, and to this cause are due the 
 asphalt pebbles of all sizes and shapes found upon the 
 beach. Where exposed to the alternate action of the sea 
 \vater and air the pitch is converted into alteration products, 
 which will be noticed later. 
 
 It is evident that the pitch forming the land deposits is, 
 as a whole, of a very varied character, and to be found in 
 all stages of alteration. 
 
 JProximate Composition and Properties of Trinidad Asphalt 
 from tJie Land Deposits. 
 
 From the results of the examinations of the specimens of 
 pitch collected in 1891 from the land deposits, the conclu- 
 sion was drawn that the relative proportions of bitumen, 
 organic matter not soluble, and inorganic or mineral matter, 
 in this form of pitch were not far different, in carefully 
 selected material from which all oxidised and altered pitch 
 had been excluded, from those found in the lake substance. 
 The amount of water, however, proved to vary somewhat in 
 the land pitch, and in the dry material the proportion of the 
 entire bitumen which was soluble in naphtha of low specific 
 gravity 0-63 to 0*64 was shown to be decidedly smaller in 
 the land than in the lake samples, while the physical 
 properties of the two kinds of asphalt also proved 
 different. 
 
 Examinations of a collection of representative land pitch 
 specimens resulted as follows. 
 
iS6 
 
 Average Composition of Lake Pitch, Dried in Vacua* 
 
 Kearney Collection. 
 
 Percent, 
 
 Bitumen- of total 
 soluble Bitumen- 
 petroleum soluble 
 naphtha. in 
 
 naphtha. 
 Per cent. 
 
 Average 
 
 Bitumen- 
 soluble 
 CS 2 . 
 
 Mineral 
 matter. 
 
 Organic 
 not 
 soluble. 
 
 Per cent. 
 
 Per cent. 
 
 Per cent. 
 
 54' 2 5 
 
 36-51 
 
 .. 9-24 
 
 65-27 
 
 Eight Specimens from Lot C, near the Lake. 
 Average .. 54*03 .. 36-49 .. 9-48 .. 33-02 .. 6i'iL 
 
 Four Specimens from Crown Land Lots Adjoining C. 
 Average .. 53 'Si .. 36-62 .. 9 '57 .. 32-29 .. 6o'OL 
 
 Five Specimens from East of Road, Middle Ground. 
 Average .. 52-31 .. 37-80 .. 9-89 .. 31-25 .. 59-74. 
 
 Seven Specimens from Village Lots, near the Gulf. 
 
 Average .. 52-27 .. 37-73 .. lo'oi .. 31 -42 .. 60*12 
 General 
 
 average.. 53-10 .. 37-16 .. 9-74 .. 31-99 .. 60-14. 
 
 These averages enable us to compare the '^composition of 
 the pitch from various parts of the deposits outside the lake, 
 among themselves, as well as with that of the lake pitch 
 analysed under similar conditions. It appears in the most 
 striking way that the further from the lake the pitch is found, 
 the more it shows signs of age, as evidenced by the increase 
 of the percentage of organic matter not soluble, that is to 
 say, of altered bitumen and of mineral matter, and,, 
 generally, a decrease in the per cent, of the total 
 bitumen, which is soluble in the naphtha used. The 
 lake pitch has 65-3 per cent, of its bitumen soluble in 
 naphtha, while just outside the lake the land pitch has only 
 6i'i, and further on only 597 per cent, soluble. This 
 
i8 7 
 
 may seem a small difference, but it is evidence of a large 
 change. In glance pitch, examined in the same way, 49 per 
 cent, only of the entire bitumen was found to be soluble in 
 naphtha, in lake prtch 65*3. Land pitch may, therefore, be 
 inferred to be about a quarter converted from lake to glance 
 pitch. 
 
 The relation between the composition of the two kinds 
 of pitch has also been shown in comparing the results of 
 analyses of some land samples. The three sets of material 
 were found to have the following average composition : 
 
 Comparative Composition of 1894 Samples. 
 
 Organic Mineral 
 Bitumen. insoluble inorganic, 
 
 Lake, aft. below surface .. 54*92 .. 9*72 .. 35*46 
 Lake boring, ijsft. deep at 
 
 centre 54-66 .. 9-44 .. 35-90 
 
 Land 52*36 ..11-24 .. 36-40 
 
 * 
 
 Bitumen Total bitumen 
 soluble in soluble in 
 
 naphtha. naphtha. 
 
 Per cent. 
 
 Lake, 2ft. below surface 31 -72 57*79 
 
 Lake boring, 1 35ft. deep at centre 31*53 57*67 
 Land 29-02 .. .. 55*43 
 
 The same relative differences are seen as in the previously- 
 mentioned series. 
 
 Again, we have the evidence of another investigator, Pro- 
 fessor S. F. Peckham, who gives, in the American Journal' 
 of Science for March, ] 896, a series of analyses of speci- 
 mens of crude lake and land pitch, which he collected 
 personally in 1895. Mr. Richardson has re-arranged these 
 analyses, classifying them according to their lake or land 
 origin, and leaving out one analysis of lake pitch, No. 21,. 
 which is plainly in error. Averages derived from these 
 
iSS 
 
 analyses, which are as follows, although carried out on 
 somewhat different lines, permit of the same conclusions in 
 regard to the changes which have taken place in the land 
 samples : 
 
 Crude Lake Pitch. 
 
 Percentage soluble in 
 
 Percentage of total 
 bitumen only soluble in 
 
 Petro- 
 leum 
 ether. 
 
 Average . . 35-2 . . 12-4 
 
 Boiling Total 
 
 spirits Chloro- bitu- 
 turpen- form. men. 
 tine. 
 
 52-8 
 
 Petro- 
 leum 
 ether. 
 
 Boiling 
 
 spirits Chloro- 
 turpen- form, 
 tine. 
 
 23-3 
 
 9-9 
 
 Crude Land Pitch. 
 Average.. 33-3 .. 11-9 .. 0-5 .. 51-7 .. 64-7 .. 23-0 .. 12-3 
 
 Iron Pitch. 
 
 No. 3 .. 33-6 .. 13-8 .. 9-9 .. 57-2 .. 58-7 
 
 24-1 .. 17-2 
 
 The determinations of bitumen made by Professor 
 Peckham with chloroform also furnish some additional and 
 conclusive evidence of the differences between the two 
 kinds of material. These show that the average land 
 specimen contains 2^4 per cent, more of its bitumen in this 
 difficultly soluble form, while iron pitch, which is acknow- 
 ledged to be of no value for paving purposes, and is always 
 rejected in digging land pitch, has 7*3 per cent, more of its 
 bitumen in this form. From this Mr. Richardson draws the 
 inference that the land pitch collected by Professor Peckham 
 is one-third converted into iron pitch, and of so much less 
 value than lake pitch for paving purposes. In fact, his results 
 are as conclusive evidence of the differences between lake and 
 land pitch as any that have been offered, and confirm the 
 results of experience with the asphalt from the land deposits 
 in the laying of street pavements. 
 
189 
 
 Composition of the Sojt Pitch and Pitch from Blow Holes- 
 
 As has been shown, the soft pitch, at the centre of the 
 lake, is in an active, state of change, so that it might readily 
 prove to be somewhat different in composition from the 
 general run of pitch from the lake. The same thing might 
 be the case with the asphalt found in a few active blow- 
 holes at or near the edge of the deposit. 
 
 Specimens of the soft asphalt have been examined, with 
 the following results the 1891 material being analysed a 
 few weeks after it was collected and while still in an active 
 state of change ; the 1893 specimens after they had ceased 
 to give off gas, and were still of rather soft consistency, and 
 easily cut with a knife : 
 
 1893. 
 
 
 1891. 
 
 1893. 
 
 Water at 100 deg. Cent 
 
 .. 
 
 (28-9) 
 
 Loss on melting 
 
 34' 1 
 
 3i-4 
 
 Bitumen in the fresh pitch . . 
 
 34-5 - 
 
 43' 
 
 Organic matter insoluble in CS.> 
 
 6-4 .. 
 
 1-9 .. 
 
 Mineral matter 
 
 25-0 .. 
 
 23-2 .. 
 
 lOO'O lOO'O 
 
 On dry substance : 
 
 Bitumen 5 2 "4 63-4 .. Co -2 
 
 Organic insoluble 9 - 6 .. 2'8 .. 6-9 
 
 Mineral matter 38*0 .. 33*8 .. 32*9 
 
 lOO'O lOO'O lOO'O 
 
 The three specimens differ considerably. In the first,, 
 the composition is similar in the dried condition to that or 
 average lake sample, having, however, more mineral matter. 
 In the two others there is less mineral and organic matter 
 not bitumen, with a consequent increase in bitumen. It 
 would seem, therefore, that the soft pitch is not as uniform 
 in composition at the source of the lake's supply as that of 
 the main mass of pitch, but by the continued movement of 
 
190 
 
 the mass becomes later of the average found in other parts 
 of the deposit. 
 
 There is also, in the case of the soft pitch, a loss of gas 
 and light oil on heating and melting which does not take 
 place with the hard asphalt to any such degree, and on 
 allowing the soft and hard pitches to flow at their softening 
 points on an inclined plane of brass, the former moves 
 5 -Sin. where the hard runs but 3*8, or 53 per cent, further. 
 This happens even after the soft pitch has been several 
 years out of the lake and becomes decidedly hardened both 
 by age and melting to drive off water. 
 
 Samples of the comparatively soft pitch from a blow-hole 
 near the power station at the edge of the lake and but a few 
 feet back from the main mass were examined to determine 
 if it was from the same source, and had the same composi- 
 tion as the lake asphalt. Following are the results : 
 
 Dried material. Per cent. 
 
 Bitumen 54 'o 
 
 Organic, not soluble 9-1 
 
 Mineral matter 36-9 
 
 100 -o 
 
 The blow-holes, of which there are two or three near the 
 lake, must be connected with the source of the main supply, 
 but they evolve a pitch unaccompanied with free water and 
 different from that welling up at the centre of the lake, 
 being more like the average lake pitch. 
 
 Water in the Pitch and in the Lake. 
 
 The analyses of the crude pitch in 1891 showed that there 
 was a loss, on heating the samples that had been brought 
 from the lake, without drying, to 100 deg. Cent., of from 
 25*8 to 30*6 per cent. This was, of course, largely water, 
 but in some cases, at the comparatively high temperature 
 
employed, some gas and light oil was driven off, while in 
 others there had been a loss of water in transit. In the 
 dried samples of 1894 it was, of course, impossible to 
 determine the water, but this has been done with carefully 
 taken samples of the crude pitch from large and fresh 
 cargoes as soon as they were discharged in New York. In 
 this way it has been found that the amount varies but little from 
 the limits, 28*5 to 29 percent. On this basis the compo- 
 sition of the crude pitch, although analysed in a dried con- 
 dition, was : 
 
 Water 28-5 
 
 Mineral matter 2 5'4 
 
 Organic, not soluble in hot carbon bisulphide . . . . 6-9 
 
 Bitumen 39-1 
 
 Of the character of the water in the crude pitch, it was 
 shown in 1892 that it was strongly mineral. Attempts to 
 collect it in any other way than by melting the crude pitch 
 and allowing it to rise to the surface have been unsuccessful, 
 although it would be more satisfactory to collect it at 
 ordinary temperatures and without condensation or change. 
 As thus secured from the still in which the fire-refined 
 asphalt is produced, and in a somewhat concentrated condi- 
 tion, it had the following characteristics : 
 
 Grms. per litre. 
 On evaporation the residue was equivalent to .. 19-872 
 
 Dried at 100 deg. Cent 17*484 
 
 130 deg. Cent 17-006 
 
 Ignited gently 15-014 
 
 These specimens of water from the stills must, of course, 
 have become somewhat concentrated beyond their normal 
 condition, but they illustrate its nature as well. It contains, 
 as most striking constituents, iodides, and borates (character- 
 istic of thermal water), and ferrous sulphate, which gives it a 
 
192 
 
 strong acid reaction. There is also a remarkably large 
 percentage of ammonia, the presence of which is of great 
 interest. The principal salts are chlorides and sulphates, 
 and its acidity is sufficiently great to attack and corrode 
 deeply the steel buckets in which the crude pitch is brought 
 from the lake to the shore. In fact, the ground and surface 
 water, wherever it comes in contact with the crude pitch, 
 acquires an acid reaction, and has been found unsuited for 
 use in boilers. 
 
 Mineral Matter in the Pitch. 
 
 The mineral matter in Trinidad asphalt consists of silica 
 principally, accompanied by some clay, oxide of iron, and 
 the substances soluble in the water of the crude pitch. 
 
 An analysis of the entire ash left on burning the crude 
 pitch resulted as follows : 
 
 S0l 
 
 add . Total. 
 
 Silica, SiO., ........ 70-64 70-64 
 
 Alumina, A1 2 O 3 ...... 7 -38 9-66 17*04 
 
 Ferric oxide, Fe 2 O.{* .. .. 6-30 1-32 7-62 
 
 Lime, CaO ........ 0-46 0-24 0-70 
 
 Magnesia, MgO ...... o-n 0-79 0-90 
 
 Soda, Na. 2 O ........ 1-56 1*56 
 
 Potassium, K. ...... '35 '35 
 
 Sulphuric oxide, SO., .... 0-97 0-97 
 
 Chlorine, Cl ..... .. .. 0-22 0-2.2 
 
 I 7'35 82-65 loc-oo 
 * FeO not determined. 
 
 The silica under the microscope appears in the form of 
 flakes with sharp fracture. It seems probable, as suggested 
 by Dr. Carl Barus and Professor Peckham, that it must 
 have been deposited from solution in the water found in the 
 pitch. The particles are naturally much larger than those 
 of the rest of the mineral matter, although in themselves 
 
193 
 
 very small. By treatment of the ash with strong acid to 
 remove the soluble portion, the silica can be obtained in a 
 white condition, but mixed with some impalpably fine white 
 clay, which cart be separated by decantation. It will be 
 seen that the insoluble portion of the mineral matter of 
 Trinidad pitch contains 85 "4 per cent, silica. 
 
 Some of the mineral matter is so impalpably fine that it 
 will not separate from a solution of melted or dried 
 Trinidad pitch in any of the usual solvents even after days 
 of standing and many hours' treatment in a centrifugal. 
 It will pass also through the finest filters. It has been 
 thought by Peckham and others, on this account, to be 
 chemically combined with the organic compounds of the 
 asphalt, but Mr. Richardson found that by continued 
 swinging in a centrifugal it can be so far reduced that it 
 amounts to but 2 per cent., and is then, apparently, only 
 in a state of mixture with the bitumen, since an analysis of 
 the very finest portion recovered by burning the pure bitu- 
 men thus swung out shows that it is a ferruginous clay, and 
 could not possibly be combined with organic matter, since 
 it consists of silicate of alumina, and a very considerable 
 amount of sulphuric acid, as well as oxide of iron. 
 
 Analysis of Finest Mineral Matter. 
 
 SiO 2 
 
 Insoluble 
 in HC1. 
 \i' 36 
 
 Soluble. 
 
 Total. 
 
 02 ' 3(> 
 
 ALO, 
 
 6-74. 
 
 3-1 6.1 
 
 
 FeXX 
 
 I ' 4.O 
 
 1 1 ' 7 J. 
 
 
 CaO 
 
 O * 4 1 ? 
 
 3" 2O 
 
 A J *4 
 
 3-6- 
 
 MO 
 
 O ' ^d. 
 
 I * AO 
 
 j u > 
 
 i -8? 
 
 K O .. 
 
 
 A ^.U 
 
 x-i8 
 
 1 5 
 i'i8 
 
 Na 2 O 
 O, 
 
 .. 
 
 o'53 
 
 7'l6 
 
 o'53 
 
 7'l6 
 
 
 
 
 
 
 41-29 
 
 58-94 
 
 100-23 
 
 
 
The Bitumen of Trinidad Asphalt. 
 
 The bitumen of Trinidad asphalt can be separated from 
 the mineral matter and organic matter not bituminous by 
 .solvents, and, as has been seen, amounts to about 39 to 40 
 per cent, of the crude pitch, or 54 to 57 per cent, of the 
 dried substance. For purposes of investigation, it is more 
 safely extracted with chloroform at a boiling temperature. 
 It is freed from suspended mineral matter by subsidation 
 and long treatment in a centrifugal machine, and from 
 traces of the solvent by heating to 200 deg. C. for a short 
 time with stirring. As thus prepared, it contains about 
 2 "3 per cent, of clay and iron oxide, which, thus far, it has 
 been impossible to remove. 
 
 The total bitumen of Trinidad asphalt is a brilliant, 
 glossy, pitch-like substance, which has a semi-conchoidal 
 fracture when struck a sharp blow, but which yields to 
 gentle pressure and slowly flows at summer temperatures. 
 It softens rapidly at 76 deg. C., and flows quickly at 83 deg. 
 C., but is not liquid until above 100 deg. C. It has a 
 specific gravity, as extracted, of 1-071 at 25 deg., and, after 
 correction for the 2 '6 per cent, of mineral matter present, 
 of 1*032. 
 
 The ultimate composition of several preparations which 
 involved complete extraction was determined with the 
 precautions necessary in burning such organic substances, 
 and gave the following results : 
 
 Total Bitumen in Trinidad Lake Asphalt. 
 
 Preparation. 
 
 I. 
 
 IV. 
 
 V. 
 
 Average. 
 
 Carbon . . 
 
 82-59 
 
 .. 81-95 
 
 .. 82-44 
 
 .. 82-33 
 
 Hydrogen 
 
 10-74 
 
 10-51 
 
 .. 10-81 
 
 .. 10-69 
 
 Sulphur 
 
 6*04 
 
 .. 6-54 
 
 5-90 
 
 .. 6-16 
 
 Nitrogen 
 
 0-51 
 
 0-92 
 
 I'OO 
 
 0-81 
 
 99-88 99 -92 100-15 99 '99 
 
195 
 
 This bitumen is characterised by the large percentage of 
 sulphur which it contains, and the presence of nitrogen. 
 There are apparently no oxygen derivatives present in the 
 bitumen, or they oecur in very minute amounts. 
 
 The Bermudez Asphalt Deposit. 
 
 From the mouth of the Orinoco, the north-eastern coast 
 of Venezuela, which faces Trinidad, is low, and consists of 
 vast mangrove swamps, through which run deep tidal 
 estuaries. That portion forming part of the State of 
 Bermudez extends inland for many miles. It lies on the 
 opposite side of the Gulf of Paria from Trinidad. About 
 thirty miles in an air line from the coast the asphalt deposit, 
 known as the Bermudez Pitch Lake, is found at the point 
 where a northern range of foot hills comes down to the 
 s\vamp. The Guanaco River, a branch of the San Juan, 
 one of the large canos or estuaries of this region, at about 
 sixty-five miles, in its winding course, from its mouth, runs 
 within three miles of the deposit, but it is five or six miles 
 to a suitable wharfage site. On the other hand, towards the 
 north a road runs to the hills and to the village of Guaryquen. 
 These are the means of communication with the deposit. 
 The so-called lake is situated between the edge of the swamp 
 and the foot hills in what might be termed a savanna. It is 
 an irregular-shaped surface, with a width of about a mile and 
 a-half from north to south and about a mile east and west. 
 Its area is a little more than 900 acres, and it is covered with 
 vegetation, high rank grass and shrubs, i ft. to 8ft. high, with 
 groves of large moriche palms, called morichales. One sees 
 no dark expanse of pitch on approaching it as at the 
 Trinidad pitch lake, and except at certain points where soft 
 pitch is welling up, nothing of the kind can be found. The 
 
 o 2 
 
196 
 
 level of the surface of the deposit does not vary more than 
 2ft., and is largely the same as that of the surrounding 
 swamps. In the rainy season it is mostly flooded, and at 
 all times very wet, so that any excavation will fill up with 
 water. These conditions make it difficult to get about 
 upon it or to excavate pitch easily. 
 
 It is readily seen that this deposit is a very different one 
 from that in the pitch lake of Trinidad. It seems to be, in 
 fact, merely an overflow of soft pitch from several springs 
 over this large expanse of savanna, and one which has not 
 the depth or uniformity of that at Trinidad. 
 
 Being on a level with the mangrove swamps, and with foot 
 hills on its other side, any large amount of asphalt could 
 hardly be held in position here, as in the old crater in 
 Trinidad, but would burst out into the swamp and be lost, 
 and, as far as borings have been made, they seem to indicate 
 but a small depth anywhere as compared with that of the 
 Trinidad lake. 
 
 At different points there is at most a depth of 7 ft. of 
 material, while the deepest part of the soft maltha is only 
 9ft., and the average of pitch below the soil and coke only 
 4ft. At points there is not more than 2ft. of pitch, and in 
 the morichales or palm groves it is often 5ft. below the 
 surface. At several points scattered over the surface are 
 areas of soft pitch, or pitch that is just exuding from springs. 
 The largest area is about 7 acres in extent and of irregular 
 shape. This has little or no vegetation upon it, and, from the 
 constant evolution of fresh pitch, is raised several feet above 
 the level of the rest of the deposit. This soft asphalt has 
 become hardened at the edges, but when exposed to the 
 sun is too soft to walk upon. The material is of the nature 
 of a maltha, and is evidently the source of all the asphalt 
 in the lake, from these exudations the pitch having spread 
 
i 9 7 
 
 in every direction, so that no great depth of pitch is found 
 even at this point. 
 
 A careful examination of the surroundings shows that in 
 one respect there is a resemblance between the point of 
 evolution of the soft pitch at the Bermudez and at the 
 Trinidad lakes. Gas is given off in considerable quantities 
 at both places, and in both cases consists partly, at least, 
 of hydrogen sulphide. At the Bermudez lake Mr. Richard- 
 son was unable to determine whether it was accompanied 
 by carbonic dioxide, but the odour of hydrogen sulphide was 
 strong. 
 
 The consistency of the soft pitch at the centre of the 
 Bermudez lake is much thinner than that of the Trinidad 
 lake. It will run like a heavy tar, and does not evolve gas 
 in the same rapid way or harden as quickly after collection. 
 It therefore does not retain the gas which is generated in it, 
 nor does the deposit as a whole do so to the same extent as 
 the Trinidad pitch. Where, however, the surface of the 
 soft pitch has toughened by exposure to the sun and air, 
 and where gas is given off beneath it, it is often raised in 
 dome-like protuberances, the beehives which were spoken of 
 by visitors to the Trinidad lake. These have a thin wall 
 of pitch, and are filled with gas which readily burns, and have 
 been seen 2ft. or more in height and i8in. in diameter. 
 They are, of course, found only near the soft spots. 
 
 Although the pitch at the Bermudez lake is too soft to 
 entangle and hold permanently the gas which is given off, 
 where the pitch of medium consistency is covered with 
 water it does not escape so readily, and thus often raises in 
 the pools of water a mushroom-like growth of pitch by the 
 reduction of the gravity of the mass from the included 
 gases. These mushrooms correspond completely, except in 
 size, with those described by Manross as existing at the 
 
198 
 
 Trinidad lake when he visited it. It seems, therefore, that 
 we have to-day several of the phenomena represented at 
 the Venezuelan lake which the hand of man has destroyed 
 at Trinidad. 
 
 There is, however, no evidence of the same simultaneous- 
 boiling up of water with the fresh soft pitch that has been 
 determined at the Trinidad lake, but that there is none at all 
 is not certain, as at the time Mr. Richardson visited the 
 locality heavy rains were falling which prevented the 
 detection of a small amount. It seems, however, improbable, 
 as the soft pitch contains little or no water, and the traces 
 found in the samples collected are probably derived from 
 rain. 
 
 Hardening of the main mass of pitch. 
 
 The soft pitch after it exudes at the centre of the 
 Bermudez lake undoubtedly hardens slowly on exposure, 
 but the condition of the surface of the main mass, which is 
 very hard and rough, and of the harder borders of the soft 
 spots, is due to other causes also. 
 
 The edges of the areas of soft asphalt are covered here 
 and there with masses of glance pitch and with black and 
 brittle cinders or coke, and which seem to have been pro- 
 duced from the maltha by fire. This is evidently the case,, 
 since the rank growth of grass, which is very dry in the dry 
 season, is particularly adapted for a rapid and intense com- 
 bustion. Such fires have been even recently started inten- 
 tionally and accidentally, and to them are due the condition 
 of the present surface of the deposit and the character of 
 much of the pitch. 
 
 The general surface of the lake is very irregular and hard. 
 There are many very narrow and irregular channels or 
 depressions from a few inches to 4ft. deep, filled with water, 
 
199 
 
 and not being easily distinguished, one often falls into them. 
 At the foot of the growth of grass and shrubs are ridges of 
 pitch mingled with soil and decayed vegetation, which have 
 been plainly coked and hardened by fires of the nature 
 which have been mentioned. When this hardened material 
 which forms only a crust is removed, asphalt of a kind suit- 
 able for paving is found. The crust is from a foot and 
 a-half to two feet in depth, and very firm, while the asphalt 
 underneath would not begin to sustain the weight which 
 that of the Trinidad pitch lake does easily. There are 
 breaks in the crust here and there through which soft pitch 
 exudes, as has been described. 
 
 It appears, therefore, that the Bermudez deposit owes its 
 existence to the exudation of a large quantity of soft maltha, 
 which is still going on, and which has spread over a great 
 extent ; that this has hardened spontaneously in the sun 
 and has also, by the action of fire, been converted over 
 almost the entire surface into a cokey crust of some depth, 
 beneath which the best material lies, and that here and 
 there are scattered masses of glance pitch produced in a 
 similar way from less violent action of heat There is no 
 evidence of a general movement and mingling of the mass 
 of this deposit in any way that would produce a uniformity 
 of composition as seen in the Trinidad pitch lake, although 
 there is a certain amount of gas evolved at the soft spots 
 where maltha exudes and some gas cavities are found in 
 the general mass of the pitch beneath the crust. 
 
 As the asphalt below the crust of the deposit is the only 
 portion of value for paving, the question has arisen as to its 
 uniformity. In order to determine this a collection of 
 samples was made at various points on the lake 
 beneath the crust, and they have been submitted to 
 analysis. 
 
200 
 
 Extremes of Determinations in the Analyses of Bermudez 
 Asphalt. 
 
 CRUDE MATERIAL. 
 
 Highest. Lowest. 
 
 Loss four days, 212 deg. Fah 46*20 .. 10-72 
 
 Loss seven hours, 400 deg. Fah 13-60 .. 4-72 
 
 DRIED MATERIAL. 
 
 Loss at 400 deg. Fah., seven hours .. 16-05 .. 5-81 
 
 Softens, deg. Fah 170 deg.. 140 deg. 
 
 Flows, deg. Fah i88deg.. 135 deg. 
 
 Mineral matter 3-65 .. 0-50 
 
 Organic not soluble 6-45 .. 0-62 
 
 Bitumen 98-52 .. 9' 6 5 
 
 Bitumen soluble naphtha 73 '5 63-40 
 
 Per cent, of total thus soluble .. .. 76-55 .. 67-78 
 
 Specific gravity 1*075 .. 1*005 
 
 Methods of Analyses. 
 
 Necessity has compelled the adoption of somewhat 
 different methods of analyses with the Bermudez samples 
 from those employed in the case of the Trinidad material, 
 since the sticky nature of this asphalt prevents its being 
 freed from water by powdering and exposure to the air. It 
 was only possible, therefore, to dry the samples by heating 
 them to such a point as to soften the bitumen and remove 
 the water, this point being kept as low as possible, but 
 undoubtedly with the loss of gas and hydrocarbons. The 
 dried samples were then used for analysis. In addition, 
 the amount of water, gas, and volatile oil lost by the 
 crude material on maintaining it for a certain length 
 of time at 212 deg. and 400 deg. Fah. was determined, 
 as well as the loss suffered by the dried material at 
 the latter temperature. The softening and flowing point 
 and the degree of softness at ordinary temperatures were 
 also determined, i softest, 2 soft, 3 medium, and 4 hard, as 
 
201 
 
 well as the specific gravity of a few samples. The deter- 
 minations of the loss at high temperatures are not as 
 accurate as the others, but with such wide variations as 
 were found, is of interest and value. 
 
 From the preceding results of analyses we see that the 
 Bermudez asphalt is quite a different material from that of 
 the Trinidad lake. It is practically free from mineral 
 matter : in fact, the softest material from the soft spot at the 
 centre, before it was in any way contaminated with extra- 
 neous matter, has no more than 0*02 per cent. It has, as 
 collected, varying amounts of water, and the loss on heating 
 to TOO deg. C. for some time may reach as high as 46 per 
 cent. Here again, however, it appears from an examination 
 of the freshly collected softest material that the water is 
 adventitious alid the loss largely light oils, as this soft 
 material contains but a trace of water. The amount of 
 organic matter insoluble is, probably, also adventitious, as 
 this is not found in the soft pitch, but it may in part be 
 derived from changes in the bitumen itself from age and 
 exposure. The portion of the total bitumen soluble in 
 naphtha is large, but not constant in the specimens from 
 the different parts of the lake, but of trie fresh material 
 nearly the entire amount is soluble. 
 
 A mere glance at the results then shows that there is a 
 great lack of uniformity in the condition and character of 
 the material found scattered over the surface of the Ber- 
 mudez deposit, and it appears that this bed of asphalt is 
 one that was originally, and is now being, derived from 
 exudations from maltha springs, scattered over the area 
 which the deposit covers ; that the original thick oil or 
 maltha was a pure bitumen free from mineral matter and 
 water which has become altered by changes within itself 
 and by external causes, until to-day it exists in the form 
 
202 
 
 found in the so-called lake. There is here, as in the 
 Trinidad pitch lake, a present source of supply which is 
 adding every year to the main body of the pitch. 
 
 Character of the Bitumen of Bermudez Asphalt. 
 
 An investigation of the nature of the bitumen of Ber- 
 mudez asphalt has been partially carried out in the same 
 way as with the Trinidad material, and has furnished some 
 results of interest. 
 
 The softest and freshest asphalt from the point where it 
 exudes has the following composition : 
 
 I. II. 
 
 Carbon 82-88 .. .. 
 
 Hydrogen 10-79 .. 
 
 Sulphur 5-87 .. .. 5.37 
 
 Nitrogen 0-75 .. .. 
 
 100-29 
 
 The striking features of this bitumen are the very high 
 percentage of sulphur, nearly as great as that in Trinidad 
 bitumen, and the notable percentage of nitrogen, while in 
 other respects it shows only evidences of being composed 
 largely of saturated and unsaturated alicyclic hydrocarbons. 
 On comparing the original bitumen with that found in the 
 general supply of the lake, as used for pavements, an in- 
 teresting change is seen. This bitumen, extracted from the 
 melted and refined pitch by chloroform, has the following 
 composition as derived from the combustion of two different 
 preparations : 
 
 Entire Bitumen Bermudez Paving Asphalt. 
 
 I. II. 
 
 Carbon . . . . . 
 
 Hydrogen . . . . 
 
 Sulphur 3-93 .. .. 3-41 
 
 Nitrogen 0-^5 .. .. 0-82 
 
20 3 
 
 Here a remarkable alteration is seen. The hardening of 
 the pitch or, very likely, the process of melting and refining 
 has removed a very considerable part of the sulphur, nearly 
 2 per cent. The sulphur must have been given off as 
 hydrogen sulphide", and in its evolution condensation has 
 taken place, and a consequent hardening of the bitumen. 
 The nitrogen remains unchanged. 
 
 Sulphur, therefore, in this case, as in that of Trinidad 
 asphalt, seems to play an important part in the hardening 
 of the bitumen. It is also found in larger amount in that 
 portion of the Bermudez bitumen which is insoluble in 
 naphtha but soluble in chloroform, as with Trinidad asphalt, 
 the bitumen soluble only in chloroform having the following 
 composition : 
 
 Per Cent. 
 
 Carbon 87*19 
 
 Hydrogen ^'47 
 
 Sulphur 4-83 
 
 Nitrogen 
 
 100-49 
 
 Sulphur in Asphalt^ and its Influence. 
 
 The interest which the presence of so much sulphur in 
 the bitumen of Trinidad and Bermudez asphalt, and 
 especially in the least soluble and harder portions, aroused 
 led me to examine a large number of asphalts and malthas 
 of varied origin, which were available in the collection of 
 the Barber Asphalt Paving Company, with a view of deter- 
 mining how general the presence of a large percentage of 
 sulphur is in bitumens, which are hard and of high melting 
 point. 
 
 At the same time complete ultimate analyses were made 
 of these bitumens, in order to show that oxygen is not to 
 
204 
 
 any extent a constituent of that portion of an asphalt 
 soluble in chloroform or bisulphide of carbon, to determine 
 how generally nitrogen is one, and what differences there 
 may be in the saturation of the hydrocarbons. 
 
 The determination of sulphur was made on the pure 
 bitumens after the method, described by Mabery, of com- 
 bustion in a current of oxygen, and absorption of the 
 sulphur dioxide formed in a standard solution of alkali. 
 In several instances, especially Trinidad pure bitumen, the 
 results were checked by determinations made by oxidation 
 with nitric acid. Mr, Richardson's experiments and those of 
 Hodgson have shown that none of the ignition methods with 
 carbonates, magnesia, or sodium peroxide give high enough 
 results.* 
 
 The analyses for carbon and hydrogen were made with 
 all the precautions which a long experience with this kind 
 of work has shown to be necessary. They were deter- 
 mined by combustion with manganese oxide Mn.,O 3 with a 
 layer of lead chromate and a copper coil on account of the 
 presence of sulphur and nitrogen. The copper coil and 
 the substance in a platinum boat were introduced into the 
 tube after the tube and its contents had been raised to a 
 red heat in current of oxygen and allowed to cool. The 
 combustion was carried on very slowly in a stream of 
 oxygen, and the absorption of the water and carbonic acid 
 carefully provided for and protected. Nitrogen was deter- 
 mined readily by the Kjeldahl method. 
 
 The results are shown on pages 206 and 207. 
 
 It appears in a general way from the preceding results 
 that all the glance pitches, or brittle asphalts, Nos. i, 3, 4, 
 and 5, contain from 976 to 8-28 per cent, of sulphur, the 
 
 \ * J. Am. Chem. Soc. 20, 882. 
 
20- 
 
 hard asphalts, not flowing at ordinary temperatures, from 
 6-47 to 3-93 per cent, while the malthas or soft asphalts 
 and heavy petroleums do not contain more than 2*29 per 
 cent. There are, however, certain exceptions. The soft 
 Bermudez and Montana asphalts should, from their con- 
 sistency, be classed with the malthas, but are placed by 
 themselves, because they are rapidly transformed under 
 certain conditions to their harder forms, which are included 
 under the asphalts, probably because of the high per- 
 centage of sulphur they contain. 
 
 The Uvalde County bitumen is also an exception, in that 
 it is of only medium hardness, although having 9-60 per 
 cent, of sulphur. This bitumen is, however, peculiar, and 
 differs from the asphalts in some of its other properties as well. 
 
 Those natural bitumens also, which are usually classed by 
 themselves, albertite, gilsonite, grahamite, wurtzelite, and 
 ozokerite, are distinguished by a low percentage of sulphur 
 and a high percentage of carbon, with the exception of 
 wurtzelite, which appears, therefore, to be more nearly 
 allied to the asphalt. They are probably, therefore, of 
 different origin from the asphalts, and as an illustration of 
 such a difference, the origin and method of formation of 
 the artificial asphalts included in the table will serve. 
 
 The so-called Pittsburg flux is produced by heating heavy 
 petroleum oil, such as Pennsylvania residuum, with sulphur. 
 Hydrogen sulphide is given off, and the oil is condensed to 
 a peculiar tough and sticky substance, but slightly affected 
 by temperature changes, and only melting at a very high 
 temperature, which has been suggested as a substitute for 
 asphalt. Here the formation of the artificial asphalt is due 
 to the condensation of olefine and paraffin hydrocarbons 
 by the action of sulphur, at a high temperature, and while 
 much is evolved, several per cent, are left in the product. 
 
206 
 
 8 8 8 8 > ' 
 a 
 
 OOOOOO^mOMOO^.OO LOO in ro 
 C -S MQMOOOOOOO 
 
 I 
 
 G 
 D 
 ^ MO>ON-^-NOOO\Tj-O^ COOCO ... 
 
 O^OOOOO^OONOOOx MpM MM 
 
 , W 
 X 
 
 <u 
 
 C OO N VO t^OO t^- O . - -- - 
 
 ^4 ONOOxONNrONrOO")! MfOM I rf ^~ 
 
 rt I>-OO t^OO 000000000000 000000 0000 ^ 
 
 I 
 
 VO O ^ N 00 t*** ^OVO O "^t* w ^00 00 fO ON ^00 
 -. C^O fO uS C4 ^j~ N M M i t POO O t^ Q\ O (*O N 
 
 en 
 
 I 
 
 rti T3 ^J T3 
 
 0^0 fcl 
 
 J ."S ~ - rt 
 
 . 
 
 -y rt ~ a; rt ^ 
 
207 
 
 o in "3- o * 
 
 
 00 in m ro 
 
 
 *=00 
 
 ' ' l '' ^ ' 
 
 
 O O O O 
 O 
 
 
 las 
 
 m in M ro 
 
 I! 1 I I 1 O ro N N 1 
 1 1 1 1 
 
 
 OO^ON_0 
 
 
 00 O 
 
 p? I . 
 
 O M M O 
 
 
 M 
 
 
 N 
 
 t~^ in ro O O 
 
 1 1 1 ro I I o -1-00 r^ m 
 
 
 yggg, 
 
 
 00 rl- N 
 
 in o w 
 
 
 
 (M OO ro in 
 
 
 t^- 
 
 
 
 
 
 
 in o n N t 1 ^ T J" 
 
 
 
 800 f> O 
 
 
 
 O Tf ro J> N * 
 
 ** 
 
 N in N 
 
 
 -O w 00 
 
 III" ' 1 ' ' in t^. t>, 
 
 00 00 00 00 00 00 
 
 1 
 
 00 00 00 00 
 
 
 aoaoao 
 
 I 
 
 
 
 
 
 
 O VO O 00 inoo t^ O N rr> O 
 N N O C>C^-<4--t-'<4-fO | - | O 
 
 I 
 
 PO C> O O 
 00 t^ 5S 
 
 *s 
 
 rf |- 
 
 
 C 
 
 in M M o 
 
 ^ 
 
 
 
 w 
 
 
 
 
 
 ^ 
 
 
 ^S 
 
 
 
 ^s 
 
 <D f .c* 
 
 *^ 
 
 X w 
 
 <ii 
 
 tq 
 
 ~ tie 
 
 nj 
 
 
 in 
 
 fcf 
 
 S "1 
 
 
 Is 
 
 
 SJ 
 
 o 
 
 
 
 
 
 ^S 
 
 
 
 
 
 ^5 
 
 * . 
 
 
 . 
 
 
 
 
 
 
 
 
 * * 
 
 
 . 
 
 " 
 
 
 
 
 
 ... g ...... 
 
 in j- 
 
 
 8| 
 
 
 " " 
 
 M : g5 : g : : : | 
 
 
 :: 1^ 
 
 
 " 
 
 ^ ^ l f^3 'rt ^ /-N ^ 
 
 
 C G 
 
 
 
 ||||| : |f|^ 
 
 
 i^3 
 
 51N 
 
 -3 " 
 
 
 S|-B 
 
 ^J 'M rQ"r^ ^ * tUj -5 ? > 
 
 
 a " 
 
 
 OJD w *^. 
 
 jl 
 
 
 jl!! 
 
 ^O<0 
 
 
 p'^ 
 
208 
 
 This substance, being formed largely from the saturated 
 paraffin hydrocarbons, is, in consequence, quite different 
 from the true asphalts in many of its properties. It will 
 not pull into strings on melting and is very short. 
 
 The sludge asphalt made from the sludge tar of the kero- 
 sene refiners is much more like natural asphalt, as it pulls 
 out to strings, like the latter melts easily, and has the 
 general physical properties of ordinary asphalts. In this 
 case these properties are due to the fact that the artificial 
 product is derived from the unsaturated hydrocarbons of 
 the petroleums, rather than the paraffins, as with Pittsburg 
 flux, and to the presence of sulphur in the reaction which 
 takes place, nearly 5 per cent, being left in the substance. 
 
 In the hard Byerlite an artificial asphalt is found which 
 is of still different origin, being produced by the action of 
 oxygen on the heavy oils of petroleum. In this way a sub- 
 stance is obtained in which sulphur is absent, and has also- 
 taken no part in the formation. It seems to bear the same 
 relation to the other artificial asphalts that gilsonite does to 
 glance pitch of asphaltic origin, and has many similar 
 physical properties. 
 
 The results of Mr. Richardson's examinations, therefore, 
 seem sufficient to show that sulphur plays an important part 
 in the hardening of asphalts, although, like the artificial 
 asphalts, some natural bitumens occur which have become 
 hardened in another way, and, perhaps, by oxygen. In 
 connection with the discussion of the nature and origin 
 of natural asphalt, these facts seem to be of considerable 
 interest. 
 
 Action of Adds on Asplialts Quantitatively Determined. 
 
 In the use of sulphuric acid of different strengths as a 
 reagent for the treatment of asphalt oils, the action was such 
 
209 
 
 that it seemed of interest to determine quantitatively, as far 
 as possible, what this action would be upon the entire 
 bitumen. These results, it seemed, would prove of relative 
 interest at least,*and probably of technical value. Of course, 
 an asphalt cannot be directly treated with acid successfully. 
 To do this satisfactorily the bitumen is dissolved iil a 
 solvent, such as naphtha, which is not strongly affected by 
 the acid. Naphtha was selected for the purpose, because, 
 although the entire bitumen of most asphalts is not soluble 
 in it, the part that is not soluble may be neglected, as it 
 would be entirely removed by the action of acid. It 
 seemed, therefore, only necessary to study the action 
 of the acid on the naphtha solution. This has been 
 done with various asphalts, and with acids of various 
 strengths, in the following manner. 
 
 A weighed amount 2 grms. of any asphalt was treated 
 with boiling 88 deg. naphtha until extracted, and then 
 diluted to 250 c.c. It was then shaken twice for exactly 
 five minutes at ordinary temperatures, with two portions of 
 the sulphuric acid. After this treatment the naphtha solu- 
 tion with its remaining bitumen was washed once with water, 
 then with alkali, and finally with water, the naphtha solution 
 being then evaporated in a weighed platinum dish, dried for 
 ten minutes at 100 deg. Cent., and the bitumen unacted 
 upon by acid determined. Following are the results, includ- 
 ing some obtained with a similar treatment of heavy 
 petroleum oils from Pennsylvania, Ohio, and Russia for 
 comparison. 
 
 It is, of course, found that the heavy petroleum oils from 
 Pennsylvania, Ohio and Russia are the least affected, as 
 they are largely made up of paraffin hydrocarbons and 
 naphthenes not readily acted on by acid in dilute naphtha 
 boluiions. Among the asphalts the malthas are least 
 
 p 
 
2JO 
 
 
 .S + - , 
 
 O fO t-* ro vO 00 
 
 1 
 
 I I s 
 
 ^t- ON iO >O rn N 00 
 
 p 
 
 s 
 
 go* 
 
 ^ONO>N m^-m 
 
 
 u^ 
 
 -1-OO 00 -^- OO O\ 
 
 3 
 
 W 
 
 
 3 
 
 . 
 
 
 9) 
 
 w O 
 
 Tl-00 ro O lOO N 
 
 ,0 
 
 5 en 
 
 -t-00 ON t^ >O r*") N 
 
 .H >. 
 
 ~S 
 
 t^O-00 CN.^Cn 
 
 J5 
 
 H o" 
 
 M ONO ^0 ^^ 
 
 .5 
 
 2 (/) 
 
 r^N. r^ r-N.oo u*i ro p^ 
 
 1 
 
 ri 
 
 O 00 00 OO O^ O^ O^ 
 
 vt-i 
 
 O 
 
 M - 
 
 oo r^co t^ 10 ^o 
 
 I 
 
 ^3 
 
 t^*OO 00 O O 1 * Q\ O 
 
 d 
 
 * 
 
 
 
 
 
 
 
 1 
 
 s- l 
 
 j_, TO 
 
 a^ 
 
 00 OO M O -^ LO O 
 
 ^- o o o r^ o o 
 
 00 00 O\ O O C^ O 
 
 c'S- 
 
 a *. 
 5 S ^ 
 
 O M O 
 
 rt^ 
 
 !|5-g 
 
 t^O O 00 O O^ O 
 
 P. rt 
 
 s 
 
 
 Ja 
 
 <u c 
 
 00 t^iO 0>00 O 
 
 
 ' C 
 
 t^ (N 10 M vO O O 
 
 I 
 
 -~4 r^ 
 
 *t-O rooo ON ON O 
 
 7 
 
 1 
 
 OMO M ON O O O 
 
 c 
 
 ! 5 
 
 4 -w 
 
 O O i^ t^. o O O 
 O ON in ON O O O 
 
 
 IS 
 
 
 
 
 
 
 
 'S ^ 
 
 
 
 
 
 
 r o ~ 
 
 
 
 a B 
 
 
 1 
 
 0; : ^6 S 
 
 C ^j ni -- 3 C ^-, 
 
 I?l8 8 8s 
 
 
 
 s iasi-i's 
 
 rt 'C ^ a; 43 O 
 
 UfflH^^Otf 
 
 p N N 
 
 OO ON ON ON 
 
 00 lO N Tf- N 
 
 O M M mvo N oo 
 10 N -^-oo ON r^ 
 
 vo co M 10 "^t* PO 
 
 8. 
 
 00 ro 10 t^00 
 
 MINN 
 
 feljll 
 
 '|-33S|g 
 
21 I 
 
 .attacked by common acid, showing that they lie between 
 the heavy petroleum oils and the asphalts, while the harder 
 the asphalt, the smaller is the per cent, of bitumen unacted 
 upon by acids, corresponding relatively to the amount 
 soluble in naphtha. 
 
 Effect of Heat on Mineral Oils. 
 
 In a recent article {Municipal Engineer ing^ August, 1897) 
 Mr. Richardson has shown that there is a difference in the 
 way different asphalts, malthas, and heavy petroleum oils are 
 .affected by being maintained at a high temperature for a 
 
 number of hours. While all of them volatilise something 
 
 -**- 
 
 -when heated to the conventional temperatures of the asphalt 
 industry 325 deg. and 400 deg. Fah. for seven hours 
 there is a very considerable difference in the appearance of 
 the residues after heating. 
 
 Petroleum residuum of the kind best suited for the 
 asphalt-paving industry, and made especially for it, may be 
 heated for seven hours at 400 deg. Fah. and lose but 2 to 
 5 per cent, of volatile oil, while the residue will still con- 
 tinue to flow at ordinary temperatures. Other petroleum 
 residuums may lose a much higher percentage of oil up to 
 15 per cent., and leave a residue, which will not flow owing 
 to the amount of scale paraffin it contains ; but in neither 
 case is the resulting residue solid and brittle. Many natural 
 heavy tarry oils, known as malthas or soft asphalts, will, 
 however, not only lose a large amount of volatile oil when 
 heated to 325 deg. or 400 deg. Fah., up to 20 or 30 per 
 cent, but they will leave behind a residue which is a brittle 
 glance pitch, showing the very different nature of the hydro- 
 carbons of which it is composed, and those making up the 
 residuum of ordinary Eastern petroleum. Malthas of a 
 medium character are also found, so that it is a very useful 
 
 p 2 
 
212 
 
 means of determining the character of any heavy mineral 
 oil to examine it in this way. 
 
 It appears that there is a very decided difference in the 
 character of the bitumens found in various asphalts, malthas,, 
 and heavy petroleum oils, which is shown in several ways by 
 the preceding methods of treatment. The amount of total 
 bitumen which is soluble in 88 deg. naphtha is, in the first 
 place, indicative of how far the action of condensation and 
 polymerisation has gone on, and of how large a part of the 
 bitumen has been converted into the hard and brittle form. 
 For example, in the heavy petroleums and malthas the 
 larger portion of the bitumen is soluble in naphtha, while in 
 the asphalts this may be reduced to 60 per cent, or less, and 
 in the glance pitches to 40. It is possible, therefore, to 
 gain considerable assistance in judging of the nature of a new 
 asphalt by such a determination, or at least, as to some of" 
 its properties of technical interest. In another way the 
 determination of the amount of the hydrocarbons soluble ini 
 naphtha, which are unacted upon by acid, helps to show the 
 nature of the substances making up this part of the bitumen. 
 And again, the stability of the hydrocarbons is shown by 
 their power of resisting high temperatures with little loss and 
 without change in their physical properties, and especially 
 their consistency. 
 
 Perhaps none of the tests which have been given possess- 
 any actual scientific merit, but they at least enable us, when 
 they are combined, to compare different asphalts, and to- 
 judge of their technical value. It must also be remembered 
 that the results have no absolute value, and can be used only 
 as a relative means of comparison, since the solvents and 
 conditions may vary from time to time. In making 
 examination of this kind, therefore, it is desirable that a well- 
 known asphalt, maltha, or oil should be subjected to a 
 
2I 3 
 
 parallel analysis, alongside of the material under investi- 
 gation. 
 
 Conclusions as to the Nature and Origin of Asphalt. 
 
 The natural bitumen, which is known as asphalt, is con> 
 posed, as far as we have been able to learn, of saturated and 
 unsaturated di-cyclic, or polycyclic, alicyclic hydrocarbons 
 and their sulphur derivatives, with a small amount of nitro- 
 genous constituents. Asphalt may therefore be defined as 
 any hard bitumen composed of such hydrocarbons and their 
 derivatives, which melts on the application of heat to a 
 viscous liquid ; while a maltha or soft asphalt may be defined 
 as a soft bitumen, consisting of alicyclic hydrocarbons, which, 
 on heating, or by other natural causes, beco'mes converted 
 into asphalt. The line between the two classes cannot 
 be sharply drawn. 
 
 Asphalts are distinguished by the large amount of sulphur 
 they contain, and it is to its presence that many of the 
 important characteristics and perhaps, in part, the origin of 
 this form of bitumen is due. The soft asphalts or malthas 
 contain much less sulphur than the harder ones, or if the 
 former are rich in sulphur, they are then in a transition stage 
 and will eventually become hard. But a small portion of 
 the constituents of a hard asphalt are volatile even in vacuo, 
 but they can be separated by solvents into an oily portion, 
 which is soft, or softens readily when heated, and a harder 
 portion, which does not melt by itself without decomposition, 
 and is a brittle solid, but soluble in the oily or softer 
 portion. The harder and least soluble portion always con- 
 tains the larger part of the sulphur. It seems, therefore, 
 that sulphur is the effectual hardening agent of natural 
 asphalts, in the same way that it is of artificial asphalts which 
 
214 
 
 are produced by heating a soft natural bitumen with 
 sulphur. 
 
 The portion of asphalt soluble only in chloroform or 
 carbon bisulphide is entirely absorbed by sulphuric acid, 
 showing that it contains no paraffins or saturated hydro- 
 carbons. 
 
 That portion which is soluble in naphtha can be, to a 
 great extent, volatilised in vacuo. The lighter constituents 
 are found to be hydrocarbons of at least two classes those 
 absorbed or converted into tars by sulphuric acid, and those 
 but slightly acted upon by it, together with sulphur com- 
 pounds partly precipitated by mercuric chloride and partly 
 recoverable from their combination with sulphuric acid by 
 steam. 
 
 The hydrocarbons absorbed by sulphuric acid and con- 
 verted into tars by it may belong to several series, less 
 saturated than the CnH.,n, and it is probable that some of 
 them or their derivatives are unsaturated polycyclic alicyclic 
 compounds. Nothing further is known of them at present. 
 Those not easily acted upon by acid are of very high 
 density and refractive index for their boiling point and 
 molecular weight, and correspond in no way with any 
 saturated or open-chain hydrocarbons. They are evidently, 
 as has been shown, saturated polycyclic hydrocarbons 
 similar to the naphthenes, but having an even greater 
 density than the latter, as far as they are known, for the 
 same molecular weight. The amount of these compara- 
 tively stable hydrocarbons unacted upon readily by sulphuric 
 acid varies in different asphalts from comparatively little to 
 30 per cent, or more. In the malthas there is a still larger 
 proportion. The greater part of an asphalt bitumen con- 
 sists, however, of unsaturated and unstable bodies liable to 
 constant change, while even the most stable constituents, 
 
215 
 
 the saturated alicyclic hydrocarbons, are slowly polymerised 
 on standing. 
 
 Origin of Asphalt. 
 
 V 
 
 Asphalt is in the process of formation to-day. It plainly 
 does not originate as such, but is a secondary product, 
 resulting from the transformation of suitable lighter forms 
 of bitumen, malthas, or even thinner oils, into harder 
 bitumen by condensation and polymerisation, a reaction in 
 which sulphur and probably sulphates seem to take an 
 important part. 
 
 Asphalt is probably derived from oils consisting of 
 alicyclic hydrocarbons, although the condensation of paraffin 
 hydrocarbons hjv sulphur or of a mixture of saturated 
 and unsaturated hydrocarbons in the same way would yield 
 similar products, as we have seen in the artificial Pittsburg 
 flux. Polymerisation would, perhaps, take place more 
 readily in the unsaturated series. 
 
 No high temperatures seem necessary in nature for the 
 change once initiated, as it is going on to-day in the 
 Trinidad pitch lake and in Venezuela. Gas is commonly 
 evolved and is largely hydrogen sulphide, the natural result 
 of the condensation of hydrocarbons by sulphur. Carbonic 
 acid accompanies it, which would seem to point to a reaction 
 between hydrocarbons and inorganic sulphates and the 
 latter as the source of the sulphur in the bitumen of asphalts. 
 This theory is based on the presence of sulphates in the 
 water emulsified with Trinidad pitch, and in the California 
 shales, while there is quite as much hydrogen sulphide given 
 off by the Bermudez asphalt which contains no water or 
 sulphates as it originates. 
 
 Whether its origin is in sulphates or inorganic sulphides, 
 or whether, as suggests itself, it is derived from the same 
 
216 
 
 substances, either animal, vegetable, or inorganic, from 
 which the hydrocarbons themselves are formed, it is plainly 
 an important factor in the production of asphalt, and 
 especially the harder ones. 
 
 It seems justifiable, therefore, to suggest that where 
 certain mineral oils, composed of alicyclic hydrocarbons, 
 originate under such circumstances as to be subjected to 
 conditions favourable to condensation and polymerisation, 
 or to the action of sulphur or sulphates, asphalt will be 
 formed, not necessarily immediately, but in the course of 
 time. 
 
 For the primary origin of asphalt one must, of course, 
 go back to that of petroleum. This subject has lately been 
 discussed at length, and both experiments and arguments 
 brought forward in favour of the animal or vegetable origin 
 of petroleum or its inorganic derivation. 
 
 The presence cf nitrogen in all the asphalts examined 
 may be considered, by those advocating the animal or 
 vegetable origin of petroleum, as confirming their views, 
 but it seems possible also to account for it from an 
 inorganic source. 
 
 The theory of the general origin of all mineral bitumen 
 in a base of unsaturated hydrocarbons suggested by 
 Heusler, seems to be supported by the unsaturated nature 
 of most of the constituents of the asphalts and malthas ; 
 but on the other hand, the formation of saturated hydro- 
 carbons from the unsaturated oils of Trinidad asphalt by 
 the action of aluminium chloride has_not yet been accom- 
 plished. 
 
 The thermal origin of the bitumen of Trinidad lake 
 asphalt is strongly suggested by the presence of a thermal 
 water, containing borates and iodides, in an emulsion of 
 definite composition with bitumen and mineral matter, 
 
217 
 
 -while the transportation of bitumen from place to place 1 y 
 means of springs is illustrated by the mineral spring which 
 rises at the centre of the Trinidad lake with the freshly 
 issuing pitch, and to the collection of bitumen in pockets, 
 even under the sea where springs break out. In no case, 
 however, have any signs of more than normal temperatures 
 been observed in any of these springs. Steam, therefore, if 
 .an element in the movement of this bitumen, must be 
 associated with it at such a point as to indicate a thermal 
 origin of the bitumen itself. 
 
 Finally, the theory of Ochsenius, that the formation of 
 petroleum is to be attributed to the action of strong salt 
 solutions on organic matter, might also be considered as 
 strengthened by the presence of such a strong brine as 
 accompanies the emulsion of soft asphalt in the Trinidad 
 lake. 
 
 Chinese and Egyptian Bitumen. Boussingault* examined 
 the bitumen from the so-called burning springs of Ho- 
 Tsing, in the Province of Szu-Tchhuan, China, where, in a 
 region having an area of fifty square miles, several thousand 
 springs exist, whence proceed inflammable gases at a high 
 velocity, accompanied by bitumen and brine. At ordinary 
 temperatures the bitumen is in a fluid condition, but when 
 cooled naphthalene separates out. The solid and liquid 
 constituents of the bitumen separated by cooling and filtra- 
 tion have the following composition : 
 
 Carbon. Hydrogen. Oxygen. Nitrogen. 
 
 Per cent. Per cent. Per cent. Per cent. 
 
 Solid 82*85 .. 13-09 .. 4-06 .. o-oo 
 
 Liquid .. .. 86-82 .. 13*16 .. o - oo .. o'oa 
 
 The same author gives the following results of the exami- 
 
 * Journal of the Society of Chemical Industry, Vol. ii 1 ., page 159. 
 
2l8 
 
 nation of samples of asphaltum respectively from Egypt and 
 from the Dead Sea : 
 
 Carbon. Hydrogen. Oxygen. Nitrogen. 
 
 Per cent. Per cent. Per cent. Per cent. 
 
 Egypt 85-29 .. 8-24 .. 6-22 .. 0-25 
 
 Dead Sea .. .. 77-84 .. 8-93 ..11-53 .. 1-70 
 
 Since the appearance of the preceding paper further 
 investigations have resulted in showing, according to Mr. 
 Richardson, that the rate of influx of asphalt into the pitch 
 lake is strongly influenced by the rapidity with which 
 material is removed from it. In two years, when the pitch 
 was taken out in larger quantities than usual, the influx 
 was greater than in ordinary years ; that is to say, the rapid 
 removal of pitch produces a more sudden removal of the 
 pressure on the source of supply, and a greater evolution of 
 fresh pitch takes place. Levels of the lake's surface are 
 still being taken yearly, and most interesting information is- 
 continually being obtained. 
 
 Investigations as to whether the finest inorganic matter, 
 not heretofore considered remcvable from the bitumen of 
 Trinidad asphalt, is actually in combination with the latter,, 
 as has been claimed, have shown that ail inorganic matter 
 can be filtered out by passing a solution of the bitumen in 
 carbon bi-sulphide through a Pasteur filter, and that orr 
 evaporation of the solvent the bitumen is quite free from 
 even a trace. The mineral matter is therefore in no way in 
 combination with the hydrocarbons of the asphalt or their 
 derivatives. 
 
 Investigations of the lighter hydrocarbons in the bitumen 
 of Trinidad asphalt and of their sulphur and nitrogen 
 derivatives have been continued, and the characteristics of 
 these substances have been compared with those of the 
 
2IQ 
 
 similar constituents of the heavy petroleums of California,, 
 and the asphalt deposits of that State being now well 
 understood, it becomes evident that the origin of the 
 bitumen in Trinidad lake asphalt is in petroleum. 
 
 THE END 
 
INDEX. 
 
 ACID, Carbonic, 50, 75 
 
 ,, Sulphuric, 52 
 Adams, Dr. M. A., 82 
 Adhesive Strength of Cement, 35 
 Adulterated Cement, 64, 104 
 Agate, 17 
 Alabaster, 6 
 
 Alkalis, Estimation of, 48 
 Allen, A., on Mortar, 79 
 Alumina, Estimation of, 45 
 
 Silicate of, 12, 17, 18 
 Amethyst, 17 
 Amorphous Silica, 17 
 Analysis of Lime, &c., 43 
 
 ,, Mortar and Cement, 
 
 53.57 
 ,, Mortar and Cement, 
 
 Typical, 66 
 Apatite, 6 
 
 Architects, Institute of British, 131 
 Arenes, French Chalk Sand, 20 
 Argillaceous Sand, 20 
 Arragonite, 6 
 Artificial Stone, 158 
 Asbestos, 17 
 Asphalt, 174 
 
 ,, Action of Acids on, 208 
 ,, Artificial, 205 
 ,, Bermudez, 195 
 
 Asphalt, Bermudez, Analysis of,2oo r 
 
 202 
 Hardening of, 
 
 198 
 . Composition of, 178, 182, 
 
 185, 206 
 
 ,, Lake, Movement of Sur- 
 face, 178 
 
 ,, Laad Depos'ts of, 184 
 ,, Methods of Analysis of, 
 
 1 80, 200 
 
 ,, Mineral Matter in, 192 
 ,, Montana, 205 
 ,, Nature and Origin of, 213 
 ,, Sulphur in, 203 
 Augite, 17 
 
 B 
 
 BACTERIA Beds, 14 
 Bad lime-mortars, 76 
 Beryl, 17 
 Beton, 145, 158 
 Bitumen, 183, 194, 210 
 
 ,, Chinese and Egyptian, 
 217 
 
 Uvalde County, 205 
 Blount, B., Stanger and, on Cement, 
 
 3, 107, 109 
 Bodmer, R., on Mortar, 79 
 
222 
 
 Eoulnois on Concrete, 156 
 Boussingault on Bitumen, 217 
 Brick Clay, 13 
 
 Piers, 35 
 
 ,, ,, Crushing Strength of, 
 141 
 
 ,, Walls, Crushing Strength of, 
 
 143 
 ,, &c., Safe Load for, 
 
 144 
 
 IBricks, Absorption of Water by, 35 
 ,, Analysis of, 70 
 ,, Crushing Strength of, 138 
 ,, Importance of Wetting, 36 
 ,, Weight of Broken, 71 
 Brick well's " Granitic-Breccia," 159, 
 
 164 
 
 Buildings Acts, 1878, Bye-laws, 25 
 Burnt Clay, 14, 71 
 
 CAERPHILLY Castle, Mortar from, 
 
 81 
 
 Calc-spar, 6 
 Calcium Carbonate, 6 
 
 ,, Phosphate, 6 
 
 Sulphate, 32 
 Cameron, Sir Chas. A., on Mortar, 
 
 78 
 Carbonic Acid, Absorption of by 
 
 Mortar, 30, 75 
 
 Carbonic Acid, Estimation of, 50 
 Cement, 9, 24, 85 
 
 ,, Artificial, 9 
 
 ,, Atkinson's, 27 
 
 ,, German, 9 
 
 ,, Keene's, 28 
 
 Cement, Medina, 27 
 ,, Medway, 9 
 ,, Mortar, 76, 88, 135, 140 
 ,, Natural, 9 
 ,, Neat, 92 
 ,, Parian, 28 
 ,, Parker's, 27 
 ,, Portland, Adulteration of, 
 
 64, 104 
 ,, ,, Analysis of, 57, 
 
 66, 112 
 ,, ,, Briquettes,Crush- 
 
 ing Strength of, 
 
 138 
 ,, ,, Constancy of 
 
 Volume, 86, 1 24 
 ,, ,, Effect of Age 
 
 on, 92 
 ,, ,, Fineness of, 27, 
 
 86, 89 
 ,, ,, for Hydraulic 
 
 Work, 92 
 ,, ,, Manufacture of, 
 
 25 
 ,, ,, Mechanical Tests 
 
 of, 94 
 ;, ,, Sand for Testing, 
 
 21 
 
 ,, ,, Specific Gravity 
 
 of, 62, 86 
 j, ,, Strength of, 21, 
 
 85, 92, 97 
 
 > M Weight of, 92 
 
 ,, ,, with Sand, 92, 93 
 
 ,, Scott's Selenitic, 27 
 Chalcedony, 17 
 Chalk, 6, 74 
 
 Chemical Analysis, 43 et seq., 60 
 1 Chinese Bitumen, 217 
 
22 3 
 
 Chipstead Valley, Flints in, 18 
 Cinders, 22, 73 
 Clay, 12, 17 
 
 ,, Analysis of, 56 
 Brick, 13 
 ,, Burnt, 14 
 China, 12 
 Fire-, 13 
 Foul, 13 
 ,, French, 18 
 Mild, 13 
 Pipe-, 13 
 Pottery, 13 
 Red, 14 
 ,, Roman, 27 
 Sandy, 13 
 Stourbridge, IJ 
 Clinker, 22 
 
 Cohesive Strength of Cement, 35 
 Colson and Colson, Messrs., Tests 
 
 by, 90 
 
 Concrete, Cement, 93, 145, 156 
 Chinese Methods, 147 
 ,, Effect of Compression on, 
 
 150 et seq. 
 
 ,, Hastening Setting of, 153 
 Historical Notes on, 145 
 in Building Construction, 
 
 148 
 ,, Mixing and Depositing, 
 
 89, 149 
 
 Roman, 147 
 Cornelian, 17 
 
 Crushing Strength of Brick Piers, 
 
 141 
 
 Walls, 
 143 
 Crystallisation Theory, 31 tt seq. 
 
 DOLOMITE, 6 
 
 EARTH, Infusorial, 17 
 Earthy Matter in Mortar, 40, 41 
 Egyptian Bitumen, 217 
 Erdmenger on Tensile Strength, 133 
 
 FAIJA'S Cement Testing Machine, 
 
 98 
 Faija on Testing Portland Cement, 
 
 94, 97 
 
 Fat Lime, 9 
 
 Felspar, 17 
 
 Fineness of Cement, 27, 86, 89 
 
 Fire-clay, 13 
 
 Fitzmaurice on Effect of Tempera- 
 ture, 153 
 
 Flint, 17, 18 
 
 ,, Microscopical Examination 
 of, 19 
 
 Fluorspar, 6 
 
 Foul Clay, 13 
 
 Fraudulent Admixtures to Cement, 
 107 et seq 
 
 French Chalk, 18 
 
 Fresenius' Cement Tests, 60 
 
 Specific Gravity Bottle, 
 60 
 
 Fuller's Earth, 12 
 
 Furnace Slag, 22, 129 
 
224 
 
 GARBUTT, Matt., on Brickwork 
 
 Tests, 136 
 
 Gasworks, Lime for, 9 
 German Cement, 9, 105 
 Glass-cubes, Action of Cement on, 
 
 35 
 Grant, J., on Cement, 26, 35, 91 
 
 ,, on Concrete, 152 
 Granolithic Stone, 165 
 Grimwood, R., 2, 3, 66 
 Grirawood, R. G., 3 
 Grit, 32, 38 
 Gypsum, 6, 28, 106, 127 
 
 HARDNESS of Water, 6 
 
 Heat, Effect on Mineral Oils, 211 
 
 Henry, P. W., on Asphalt Borings, 
 
 175 
 
 Highton's Victoria Stone, 159 
 Hilton and Co , Fineness of Cement, 
 
 27 
 ,, ,, Percolation Test, 
 
 1 02 
 
 Hornblende, 17 
 Hughes, J., on Mortar, 80 
 Hydraulic Lime, 9 
 
 INFUSORIAL Earth, 17 
 Iron, Estimation of, 44, 45 
 
 JADE, 18 
 Jasper, 17 
 
 KAOLIN, 12 
 
 Keates' Specific Gravity Bottle, 
 
 Kentish Ragstone, 104 
 
 LAPIS-LAZUI.I, 18 
 
 Landrin, E., on Absorption of 
 
 Lime by Silica, 155 
 Legal Specification of Mortar, 67 
 Leighton Buzzard Sand, 21, 138 
 Lesley on TensiJe Strength, 91 
 Lewis, F. H., on Cement, 85 
 Lime, 5, 43 
 
 r , Analysis of, 46, 70 
 
 ,, Calculation of Value of, 1 1 
 
 Fat, 9 
 
 ,, for Gasworks, 9 
 
 ,, Hydraulic, 9 
 
 Percentage of Pure, II 
 
 Poor, 9 
 
 Quick-, 7 
 
 Slaked, 7 
 
 Specification for, 10 
 
 Briquettes, Crushing Strength 
 of, 138 
 
 ., and Cement Mortars, Analy- 
 sis of, 71, 72, 140 
 
 Mortars, 24, 37, 53, 67, 71,. 
 
 72, 76, 78 
 
 Limestone, 6, 43, 72 
 Lime Composition, 8 
 Lime Water, 5 
 Loam, 12, 13 
 
 London Chamber of Commerce, 106 
 Lundteigen on Tensile Strength 
 
 133 
 
22 5 
 
 M 
 
 MACKAY, W. W., on Temperature, 
 
 153 
 Magnesia, Estimation of, 47 
 
 ,, in Limestone, 8 
 ,, Silicate of, 17, 1 8 
 
 Malm, 14 
 
 Maltha, 199, 207 
 
 Marble, 6 
 
 ,, Moreau, 166 
 
 Marl, 14 
 
 Mechanical Tests of Cement, 85,94 
 
 Medway Cement, 9 
 
 Meerschaum, ij 
 
 Mica, 1 8 
 
 Microscopical Examination of 
 Cement, 121 
 
 Microscopical Examination of 
 Quartz and Flint, 19 
 
 Mild Clay, 13 
 
 Mineral Oils, Effect of Heat on, 
 
 211 
 
 Moisture in Lime, 44 
 Moreau Marble, 166 
 Mortar, 24, yfot scq 
 
 Analysis of, 53, 71, ^^ 
 
 Bad, 76 
 
 Cement, 76, 88, 89 
 
 Composition of, 25, 68 
 
 Good, 78 
 
 Grit in, 38 
 
 Legal Specification of, 67 
 
 Mixing, 88 
 
 Technical Examination of, 
 
 37 
 
 Setting of, 24 
 Vegetable debris in, 38 
 
 N 
 
 NEWLANTS, B. E. R., on Mor(ar, 82 
 Non-slip Stone, 165 
 
 OLAVINE, 18 
 Opal, 17 
 
 PARKER'S Cement, 27 
 Piers, Brick, 35 
 Pipe-clay, 14 
 Pit Sand, 19 
 Pitch (see Asphalt), 174 
 Pittsburg Flux, 205 
 Plaster of Paris, 28, 32 
 Poor Lime, 9 
 Portland Cement, 
 
 ,, j, Adulteration of, 
 
 104 
 
 ,, ,, Analysis of, 57 
 
 ,, ,, Boiling Test, 91 
 
 ,, ,, Briquettes, Crush- 
 
 ing Strength of, 
 138 
 
 Concrete, 93 
 ,, Cost of Testing, 
 
 87 
 
 ,, ,, Constancy of 
 
 volume, 86, 124 
 
 ,, ,, Effect of tempe- 
 
 rature on, 153 
 
 ,, Fineness, 27 
 
 Gauging, 95 
 
 Q 
 
226 
 
 Portland Cement, Gypsum in, 6, 28, 
 
 106, 127, 128 
 Manufacture, 25 
 ,, ,, Sampling, 94 
 
 Slag in, 64, 107, 
 
 129 
 ,, Specification of, 
 
 96 
 ,, Tensile Strength, 
 
 97 
 
 ,, Testing, 94 
 Pottery clay, 13 
 Pouzzo-Portland, 155 
 Pumice, 18 
 Puscher, E., on Hardening Cement, 
 
 154 
 Puzzuolana, 22 
 
 Analysis of, 23 
 
 QUARTZ, 17, 18, 19 
 
 ,, Ferruginous, 18 
 Quick-lime, 7 
 Quick-setting Portland Cement, 87 
 
 RAMMING Concrete, 152 
 Ranger's Artificial Concrete, 158 
 Ransome's Siliceous Stone, 158 
 Red Clay, 14 
 Richardson, Clifford, on Asphalt, 
 
 1 74 et se<] 
 River Sand, 19 
 
 Rochester Castle, Mortar from, Si 
 Roman cement, 94 
 ,, mortar, 22 
 
 SALT Water, Cement for, 87, 93 
 Salter and Co.'s Cement Testing 
 
 Machine, 101 
 Salting-out, 7, 19 
 Sand, 17, 19, 34, 88 
 
 ,, Argillaceous, 20 
 
 ,, Fineness of, 21 
 
 ,, Leighton Buzzard, 21, 138 
 
 ,, Microscopical Examination 
 of, 19 
 
 Pit, 19, 73 
 
 River, 19, 73 
 
 Sea, 19 
 
 Sharp, 19 
 
 Standard, 21 
 
 ,, Substitutes for, 21 
 
 Thames, 21 
 
 ,, Washing, 21 
 
 ,, Water-worn, 19 
 Sandstone, 17 
 c andy Clay, 13 
 Schist, 20 
 Scoriae, 22 
 
 Searles-Wood, Mr., 3 
 Selenite, 6 
 Setting of Mortars and Cements, 
 
 30, 40, 130 
 Shale, 13 
 Siderite, 18 
 Silica, 17, 31,44, 155 
 
 ,, Amorphous, 17 
 
 Soluble, 69 
 Silicate of Alumina, 12, 17, 18 
 
 ,, Magnesia, 17, 1 8 
 Siliceous Stone, 158 
 Slag, Furnace, 22 
 
 in Cement, 64, 107, 129 
 Slaked Lime, 7 
 
227 
 
 Slate, i 8 
 
 Slow-setting Portland Cement, 88 
 Slurry, 9 
 
 Smith, C. Chambers, 3, 73 
 Soap Stone, 18 
 
 Sorby, Dr., on Flint and Quartz, 19 
 Specific Gravity Bottle, Keates', 58 
 Fresenius, 60 
 
 of Portland Cement, 
 
 62,86 
 Specification for Concrete, 156 
 
 Lime, 10 
 
 Stanger and Blount on Cement, 2, 
 
 107, 109 ct seq 
 Steatite, 18 
 
 Stone, Artificial, 15** et seq 
 Stone Lime, Analysis of, 72 
 Stourbridge Clay, 13 
 Strength of Cement, 21 
 
 Brickwork, 134, 138 
 
 Mortar, 38 
 
 Piers and Walls, 141, 143 
 
 Stuart's Granolithic Stone, 165 
 Stucco, 29 
 
 Sulphur in Asphalt, 203, 208 
 Sulphuric Acid, Estimation of, 52 
 
 TABULATION of Tests. 63 
 Talc, 1 8 
 
 Technical Te.sts of Mortar, 39 
 Temperature, Effect of on Cement. 
 153 
 
 Tensile Strength, 91', 92, 110, 133 
 Testing Cement, Grant on. 94 
 Sand for, 21 
 
 Machines, Faija's, 98 
 Salter and Co.'s, 
 
 101 
 
 Terra-cotta, 169 
 
 Thames Sand, 21 
 
 Tintern Abbey, Mortar from. 81 
 
 Topaz, 1 8 
 
 Trass, Analysis of, 23 
 
 Trinidad Asphalt, 174 
 
 VALUE of Lime, n 
 Vegetable dibris in Mortar; 38 
 Victoria Stone, 159 
 
 W 
 
 WARD'S Stone, 164 
 
 \Vater, Effect on Bricks and Cement. 
 
 93 
 
 for Cement, 87, 88, 93, 132 
 
 Hardness of, 6 
 
 Lime, 5 
 
 Softening, 31 
 
 Worn Sand, 19 
 Weights of Materials, 73 
 Weight of Portland Cement. 92 
 
RETURN TO the circulation desk of any 
 
 University of California Library 
 
 or to the 
 
 NORTHERN REGIONAL LIBRARY FACILITY 
 Bldg. 400, Richmond Field Station 
 University of California 
 Richmond, CA 94804-4698 
 
 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS 
 2- month loans may be renewed by calling 
 
 (510)642-6753 
 1-year loans may be recharged by bringing books 
 
 to NRLF 
 Renewals and recharges may be made 4 days 
 
 prior to due date 
 
 DUE AS STAMPED BELOW 
 
 FEB 6 1996 
 
 IAN a 01996 
 
 20,000 (4/94) 
 
104575 
 
 YB I 1099 
 
 
 ' . - & 
 
 m